e9545c4dde
Model: fpadovani/dan-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
10145 lines
276 KiB
JSON
10145 lines
276 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.184100418410042,
|
|
"eval_steps": 500,
|
|
"global_step": 5000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 5.703905439376831,
|
|
"epoch": 1.6778242677824267,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004907298562603431,
|
|
"loss": 5.4048,
|
|
"mean_token_accuracy": 0.18117905259132386,
|
|
"num_tokens": 3711022.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.7350818634033205,
|
|
"epoch": 1.6820083682008367,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004906379415245896,
|
|
"loss": 5.5196,
|
|
"mean_token_accuracy": 0.1730467975139618,
|
|
"num_tokens": 3719778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.668873691558838,
|
|
"epoch": 1.6861924686192469,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004905455830406215,
|
|
"loss": 5.5049,
|
|
"mean_token_accuracy": 0.1729239508509636,
|
|
"num_tokens": 3728773.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.716511297225952,
|
|
"epoch": 1.6903765690376569,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004904527809984978,
|
|
"loss": 5.5406,
|
|
"mean_token_accuracy": 0.17308518290519714,
|
|
"num_tokens": 3737781.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.861098527908325,
|
|
"epoch": 1.694560669456067,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004903595355891902,
|
|
"loss": 5.5839,
|
|
"mean_token_accuracy": 0.16120259910821916,
|
|
"num_tokens": 3748010.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.702961492538452,
|
|
"epoch": 1.698744769874477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004902658470045826,
|
|
"loss": 5.555,
|
|
"mean_token_accuracy": 0.1690961465239525,
|
|
"num_tokens": 3757894.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.821532773971557,
|
|
"epoch": 1.702928870292887,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004901717154374713,
|
|
"loss": 5.6722,
|
|
"mean_token_accuracy": 0.15884690433740617,
|
|
"num_tokens": 3768760.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.750804805755616,
|
|
"epoch": 1.707112970711297,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004900771410815638,
|
|
"loss": 5.4919,
|
|
"mean_token_accuracy": 0.170400907099247,
|
|
"num_tokens": 3778273.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.63422269821167,
|
|
"epoch": 1.711297071129707,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004899821241314788,
|
|
"loss": 5.5026,
|
|
"mean_token_accuracy": 0.17012445330619813,
|
|
"num_tokens": 3787012.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.758808755874634,
|
|
"epoch": 1.715481171548117,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000489886664782746,
|
|
"loss": 5.4574,
|
|
"mean_token_accuracy": 0.17343221306800843,
|
|
"num_tokens": 3795144.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.84836859703064,
|
|
"epoch": 1.719665271966527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004897907632318055,
|
|
"loss": 5.6585,
|
|
"mean_token_accuracy": 0.1640017732977867,
|
|
"num_tokens": 3804534.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.7448265075683596,
|
|
"epoch": 1.723849372384937,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004896944196760071,
|
|
"loss": 5.4484,
|
|
"mean_token_accuracy": 0.17192598283290864,
|
|
"num_tokens": 3813757.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.620305347442627,
|
|
"epoch": 1.7280334728033473,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895976343136106,
|
|
"loss": 5.4847,
|
|
"mean_token_accuracy": 0.1775953769683838,
|
|
"num_tokens": 3823157.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.63016095161438,
|
|
"epoch": 1.7322175732217573,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895004073437844,
|
|
"loss": 5.3719,
|
|
"mean_token_accuracy": 0.17575036585330964,
|
|
"num_tokens": 3831661.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.613941764831543,
|
|
"epoch": 1.7364016736401675,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004894027389666061,
|
|
"loss": 5.3794,
|
|
"mean_token_accuracy": 0.176898692548275,
|
|
"num_tokens": 3840591.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.762122535705567,
|
|
"epoch": 1.7405857740585775,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004893046293830616,
|
|
"loss": 5.4919,
|
|
"mean_token_accuracy": 0.1701211228966713,
|
|
"num_tokens": 3850094.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.690495014190674,
|
|
"epoch": 1.7447698744769875,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004892060787950447,
|
|
"loss": 5.5638,
|
|
"mean_token_accuracy": 0.1685883015394211,
|
|
"num_tokens": 3858973.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.599520683288574,
|
|
"epoch": 1.7489539748953975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004891070874053564,
|
|
"loss": 5.3319,
|
|
"mean_token_accuracy": 0.17937298715114594,
|
|
"num_tokens": 3868315.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.69226450920105,
|
|
"epoch": 1.7531380753138075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004890076554177051,
|
|
"loss": 5.5003,
|
|
"mean_token_accuracy": 0.16688059717416764,
|
|
"num_tokens": 3877605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.723655652999878,
|
|
"epoch": 1.7573221757322175,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004889077830367061,
|
|
"loss": 5.4986,
|
|
"mean_token_accuracy": 0.17630131989717485,
|
|
"num_tokens": 3886998.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.730270290374756,
|
|
"epoch": 1.7615062761506275,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004888074704678806,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.17330997288227082,
|
|
"num_tokens": 3896018.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.7284965991973875,
|
|
"epoch": 1.7656903765690377,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004887067179176558,
|
|
"loss": 5.4595,
|
|
"mean_token_accuracy": 0.17687606066465378,
|
|
"num_tokens": 3904957.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.688749885559082,
|
|
"epoch": 1.7698744769874477,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004886055255933639,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.16950775533914567,
|
|
"num_tokens": 3913427.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.7210675239562985,
|
|
"epoch": 1.7740585774058577,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004885038937032429,
|
|
"loss": 5.4966,
|
|
"mean_token_accuracy": 0.17055633962154387,
|
|
"num_tokens": 3923134.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.680647325515747,
|
|
"epoch": 1.778242677824268,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004884018224564345,
|
|
"loss": 5.3977,
|
|
"mean_token_accuracy": 0.1748950868844986,
|
|
"num_tokens": 3931867.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.712876415252685,
|
|
"epoch": 1.782426778242678,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004882993120629852,
|
|
"loss": 5.4979,
|
|
"mean_token_accuracy": 0.17150880992412568,
|
|
"num_tokens": 3941057.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.8233030319213865,
|
|
"epoch": 1.786610878661088,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004881963627338449,
|
|
"loss": 5.6191,
|
|
"mean_token_accuracy": 0.16467538326978684,
|
|
"num_tokens": 3950511.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.720336484909057,
|
|
"epoch": 1.790794979079498,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048809297468086654,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.17224000692367553,
|
|
"num_tokens": 3959769.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.723703384399414,
|
|
"epoch": 1.794979079497908,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004879891481168063,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.17321595996618272,
|
|
"num_tokens": 3968462.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.650881338119507,
|
|
"epoch": 1.799163179916318,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048788488325532235,
|
|
"loss": 5.4878,
|
|
"mean_token_accuracy": 0.1740339145064354,
|
|
"num_tokens": 3977464.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.653697538375854,
|
|
"epoch": 1.803347280334728,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004877801803109753,
|
|
"loss": 5.4079,
|
|
"mean_token_accuracy": 0.17800928503274918,
|
|
"num_tokens": 3986303.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.739394474029541,
|
|
"epoch": 1.8075313807531381,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048767503949922683,
|
|
"loss": 5.4614,
|
|
"mean_token_accuracy": 0.17407658249139785,
|
|
"num_tokens": 3994970.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.767380714416504,
|
|
"epoch": 1.8117154811715481,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004875694610364398,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.17009811997413635,
|
|
"num_tokens": 4004679.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.775309467315674,
|
|
"epoch": 1.8158995815899581,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048746344513987775,
|
|
"loss": 5.5612,
|
|
"mean_token_accuracy": 0.16578367352485657,
|
|
"num_tokens": 4013516.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.7162766456604,
|
|
"epoch": 1.8200836820083683,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004873569920277043,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.1787016436457634,
|
|
"num_tokens": 4022335.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.582120418548584,
|
|
"epoch": 1.8242677824267783,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048725010191898296,
|
|
"loss": 5.4052,
|
|
"mean_token_accuracy": 0.17675473988056184,
|
|
"num_tokens": 4031295.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.763207721710205,
|
|
"epoch": 1.8284518828451883,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048714277503367627,
|
|
"loss": 5.625,
|
|
"mean_token_accuracy": 0.1708970382809639,
|
|
"num_tokens": 4040378.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.807916355133057,
|
|
"epoch": 1.8326359832635983,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048703501159264576,
|
|
"loss": 5.4916,
|
|
"mean_token_accuracy": 0.16382087618112565,
|
|
"num_tokens": 4050217.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.575899696350097,
|
|
"epoch": 1.8368200836820083,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004869268118176512,
|
|
"loss": 5.3676,
|
|
"mean_token_accuracy": 0.17357392609119415,
|
|
"num_tokens": 4059088.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.743972587585449,
|
|
"epoch": 1.8410041841004183,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048681817593135046,
|
|
"loss": 5.4853,
|
|
"mean_token_accuracy": 0.1811179280281067,
|
|
"num_tokens": 4067609.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.730962657928467,
|
|
"epoch": 1.8451882845188283,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048670910415729855,
|
|
"loss": 5.462,
|
|
"mean_token_accuracy": 0.1765058621764183,
|
|
"num_tokens": 4076763.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.732498407363892,
|
|
"epoch": 1.8493723849372385,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048659959671994786,
|
|
"loss": 5.6073,
|
|
"mean_token_accuracy": 0.1676172599196434,
|
|
"num_tokens": 4086552.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.614617490768433,
|
|
"epoch": 1.8535564853556485,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048648965384464693,
|
|
"loss": 5.3896,
|
|
"mean_token_accuracy": 0.1742885783314705,
|
|
"num_tokens": 4096200.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.616799926757812,
|
|
"epoch": 1.8577405857740585,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048637927575764057,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.17480439245700835,
|
|
"num_tokens": 4105080.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.676580953598022,
|
|
"epoch": 1.8619246861924688,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048626846268606913,
|
|
"loss": 5.3657,
|
|
"mean_token_accuracy": 0.17552304714918138,
|
|
"num_tokens": 4113972.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.739257526397705,
|
|
"epoch": 1.8661087866108788,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048615721485796803,
|
|
"loss": 5.6047,
|
|
"mean_token_accuracy": 0.16363925933837892,
|
|
"num_tokens": 4123615.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.715078544616699,
|
|
"epoch": 1.8702928870292888,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004860455325022676,
|
|
"loss": 5.3897,
|
|
"mean_token_accuracy": 0.17537097334861756,
|
|
"num_tokens": 4132128.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.765444278717041,
|
|
"epoch": 1.8744769874476988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000485933415848792,
|
|
"loss": 5.5964,
|
|
"mean_token_accuracy": 0.16871390193700792,
|
|
"num_tokens": 4141575.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.713302230834961,
|
|
"epoch": 1.8786610878661087,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004858208651282593,
|
|
"loss": 5.4477,
|
|
"mean_token_accuracy": 0.17084982097148896,
|
|
"num_tokens": 4150372.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.6972448348999025,
|
|
"epoch": 1.8828451882845187,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000485707880572281,
|
|
"loss": 5.5379,
|
|
"mean_token_accuracy": 0.1768944799900055,
|
|
"num_tokens": 4159007.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.661499166488648,
|
|
"epoch": 1.8870292887029287,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004855944624133608,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.1670408234000206,
|
|
"num_tokens": 4168646.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.680505037307739,
|
|
"epoch": 1.891213389121339,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048548061088489523,
|
|
"loss": 5.4848,
|
|
"mean_token_accuracy": 0.16732101887464523,
|
|
"num_tokens": 4177979.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.712616205215454,
|
|
"epoch": 1.895397489539749,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048536632622117246,
|
|
"loss": 5.4185,
|
|
"mean_token_accuracy": 0.17556924372911453,
|
|
"num_tokens": 4186781.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.640240478515625,
|
|
"epoch": 1.899581589958159,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048525160865737186,
|
|
"loss": 5.4755,
|
|
"mean_token_accuracy": 0.17089450508356094,
|
|
"num_tokens": 4195661.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.6722348690032955,
|
|
"epoch": 1.9037656903765692,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048513645842956384,
|
|
"loss": 5.3834,
|
|
"mean_token_accuracy": 0.1783241108059883,
|
|
"num_tokens": 4203902.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.627431011199951,
|
|
"epoch": 1.9079497907949792,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048502087577470895,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.16713648438453674,
|
|
"num_tokens": 4213663.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.715233516693115,
|
|
"epoch": 1.9121338912133892,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048490486093065774,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.16914888471364975,
|
|
"num_tokens": 4222152.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.687997198104858,
|
|
"epoch": 1.9163179916317992,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048478841413615017,
|
|
"loss": 5.4996,
|
|
"mean_token_accuracy": 0.17492978423833846,
|
|
"num_tokens": 4232347.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.698093748092651,
|
|
"epoch": 1.9205020920502092,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048467153563081487,
|
|
"loss": 5.5576,
|
|
"mean_token_accuracy": 0.17761845588684083,
|
|
"num_tokens": 4240607.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.755454874038696,
|
|
"epoch": 1.9246861924686192,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004845542256551692,
|
|
"loss": 5.5519,
|
|
"mean_token_accuracy": 0.17238390892744065,
|
|
"num_tokens": 4251131.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.694919157028198,
|
|
"epoch": 1.9288702928870292,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004844364844506181,
|
|
"loss": 5.5113,
|
|
"mean_token_accuracy": 0.17442891895771026,
|
|
"num_tokens": 4259955.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.772314262390137,
|
|
"epoch": 1.9330543933054394,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004843183122594541,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.17702126652002334,
|
|
"num_tokens": 4268686.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.655993032455444,
|
|
"epoch": 1.9372384937238494,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004841997093248566,
|
|
"loss": 5.4648,
|
|
"mean_token_accuracy": 0.17248067855834961,
|
|
"num_tokens": 4277377.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.60088005065918,
|
|
"epoch": 1.9414225941422594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004840806758908913,
|
|
"loss": 5.4408,
|
|
"mean_token_accuracy": 0.17208822220563888,
|
|
"num_tokens": 4287037.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.799388265609741,
|
|
"epoch": 1.9456066945606696,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048396121220250983,
|
|
"loss": 5.5864,
|
|
"mean_token_accuracy": 0.16356464326381684,
|
|
"num_tokens": 4296923.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.623253631591797,
|
|
"epoch": 1.9497907949790796,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004838413185055494,
|
|
"loss": 5.3509,
|
|
"mean_token_accuracy": 0.180958990752697,
|
|
"num_tokens": 4305873.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.6483683586120605,
|
|
"epoch": 1.9539748953974896,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004837209950467318,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.17217080146074296,
|
|
"num_tokens": 4314912.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.6739672183990475,
|
|
"epoch": 1.9581589958158996,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004836002420736635,
|
|
"loss": 5.4958,
|
|
"mean_token_accuracy": 0.16771866679191588,
|
|
"num_tokens": 4324606.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.65278697013855,
|
|
"epoch": 1.9623430962343096,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004834790598348347,
|
|
"loss": 5.3821,
|
|
"mean_token_accuracy": 0.1769419178366661,
|
|
"num_tokens": 4333076.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.695239591598511,
|
|
"epoch": 1.9665271966527196,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004833574485796188,
|
|
"loss": 5.5105,
|
|
"mean_token_accuracy": 0.17234102189540862,
|
|
"num_tokens": 4341893.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.6243753910064695,
|
|
"epoch": 1.9707112970711296,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048323540855827244,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.16903669238090516,
|
|
"num_tokens": 4352303.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.743954133987427,
|
|
"epoch": 1.9748953974895398,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004831129400219341,
|
|
"loss": 5.4403,
|
|
"mean_token_accuracy": 0.17122059613466262,
|
|
"num_tokens": 4361166.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.708210897445679,
|
|
"epoch": 1.9790794979079498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048299004322262457,
|
|
"loss": 5.4863,
|
|
"mean_token_accuracy": 0.1703219473361969,
|
|
"num_tokens": 4369999.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.668295049667359,
|
|
"epoch": 1.9832635983263598,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048286671841324565,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.174606616795063,
|
|
"num_tokens": 4379373.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.647440624237061,
|
|
"epoch": 1.98744769874477,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004827429658475799,
|
|
"loss": 5.4502,
|
|
"mean_token_accuracy": 0.17817381918430328,
|
|
"num_tokens": 4389035.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.706187343597412,
|
|
"epoch": 1.99163179916318,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048261878578029026,
|
|
"loss": 5.6031,
|
|
"mean_token_accuracy": 0.16882529556751252,
|
|
"num_tokens": 4399366.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.688080549240112,
|
|
"epoch": 1.99581589958159,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004824941784669193,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.17299768924713135,
|
|
"num_tokens": 4408183.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.665106153488159,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004823691441638889,
|
|
"loss": 5.5564,
|
|
"mean_token_accuracy": 0.16316515654325486,
|
|
"num_tokens": 4418644.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.608956384658813,
|
|
"epoch": 2.00418410041841,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004822436831284995,
|
|
"loss": 5.1876,
|
|
"mean_token_accuracy": 0.1905205264687538,
|
|
"num_tokens": 4427828.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.603258466720581,
|
|
"epoch": 2.00836820083682,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004821177956189297,
|
|
"loss": 5.1059,
|
|
"mean_token_accuracy": 0.188554809987545,
|
|
"num_tokens": 4436194.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.558306360244751,
|
|
"epoch": 2.01255230125523,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004819914818942359,
|
|
"loss": 5.1727,
|
|
"mean_token_accuracy": 0.18965528756380082,
|
|
"num_tokens": 4444468.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.661289262771606,
|
|
"epoch": 2.01673640167364,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048186474221435127,
|
|
"loss": 5.095,
|
|
"mean_token_accuracy": 0.19564822763204576,
|
|
"num_tokens": 4452760.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.600844764709473,
|
|
"epoch": 2.02092050209205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048173757684008583,
|
|
"loss": 5.1545,
|
|
"mean_token_accuracy": 0.19198492020368577,
|
|
"num_tokens": 4462154.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.612996435165405,
|
|
"epoch": 2.0251046025104604,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004816099860331255,
|
|
"loss": 5.2402,
|
|
"mean_token_accuracy": 0.18920202255249025,
|
|
"num_tokens": 4471987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.559755516052246,
|
|
"epoch": 2.0292887029288704,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048148197005603145,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.18673418015241622,
|
|
"num_tokens": 4481645.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.531320238113404,
|
|
"epoch": 2.0334728033472804,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048135352917224005,
|
|
"loss": 5.1473,
|
|
"mean_token_accuracy": 0.19193678200244904,
|
|
"num_tokens": 4491166.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.6570638656616214,
|
|
"epoch": 2.0376569037656904,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048122466364606214,
|
|
"loss": 5.1568,
|
|
"mean_token_accuracy": 0.18762783706188202,
|
|
"num_tokens": 4500159.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.490546846389771,
|
|
"epoch": 2.0418410041841004,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048109537374268216,
|
|
"loss": 5.0973,
|
|
"mean_token_accuracy": 0.2007716864347458,
|
|
"num_tokens": 4509737.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.37498288154602,
|
|
"epoch": 2.0460251046025104,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048096565972815793,
|
|
"loss": 5.0333,
|
|
"mean_token_accuracy": 0.19452138394117355,
|
|
"num_tokens": 4518379.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.5674927711486815,
|
|
"epoch": 2.0502092050209204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004808355218694199,
|
|
"loss": 5.166,
|
|
"mean_token_accuracy": 0.1909705013036728,
|
|
"num_tokens": 4527885.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.628273820877075,
|
|
"epoch": 2.0543933054393304,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048070496043427116,
|
|
"loss": 5.2023,
|
|
"mean_token_accuracy": 0.19097182154655457,
|
|
"num_tokens": 4537894.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.581399822235108,
|
|
"epoch": 2.0585774058577404,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048057397569138596,
|
|
"loss": 5.0974,
|
|
"mean_token_accuracy": 0.19794044345617295,
|
|
"num_tokens": 4546462.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.516513252258301,
|
|
"epoch": 2.062761506276151,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004804425679103099,
|
|
"loss": 5.1394,
|
|
"mean_token_accuracy": 0.18676456511020662,
|
|
"num_tokens": 4555970.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.5772850036621096,
|
|
"epoch": 2.066945606694561,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004803107373614589,
|
|
"loss": 5.1936,
|
|
"mean_token_accuracy": 0.1856154203414917,
|
|
"num_tokens": 4565799.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.580361700057983,
|
|
"epoch": 2.071129707112971,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004801784843161194,
|
|
"loss": 5.2199,
|
|
"mean_token_accuracy": 0.17868874371051788,
|
|
"num_tokens": 4575571.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.585469436645508,
|
|
"epoch": 2.075313807531381,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004800458090464465,
|
|
"loss": 5.128,
|
|
"mean_token_accuracy": 0.19369832873344422,
|
|
"num_tokens": 4584263.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.514052104949951,
|
|
"epoch": 2.079497907949791,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047991271182546476,
|
|
"loss": 5.0915,
|
|
"mean_token_accuracy": 0.19910650551319123,
|
|
"num_tokens": 4592979.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.508087635040283,
|
|
"epoch": 2.083682008368201,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047977919292706696,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.19578083008527755,
|
|
"num_tokens": 4602042.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.544242286682129,
|
|
"epoch": 2.087866108786611,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004796452526260134,
|
|
"loss": 5.1783,
|
|
"mean_token_accuracy": 0.18945651948451997,
|
|
"num_tokens": 4611197.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.44640588760376,
|
|
"epoch": 2.092050209205021,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047951089119793164,
|
|
"loss": 5.1287,
|
|
"mean_token_accuracy": 0.18829671293497086,
|
|
"num_tokens": 4621612.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.092050209205021,
|
|
"eval_entropy": 5.304144553124051,
|
|
"eval_loss": 5.5294976234436035,
|
|
"eval_mean_token_accuracy": 0.18056033307911185,
|
|
"eval_num_tokens": 4621612.0,
|
|
"eval_runtime": 2.0958,
|
|
"eval_samples_per_second": 1864.184,
|
|
"eval_steps_per_second": 233.321,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.441925144195556,
|
|
"epoch": 2.096234309623431,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000479376108919316,
|
|
"loss": 5.0075,
|
|
"mean_token_accuracy": 0.20454216599464417,
|
|
"num_tokens": 4630198.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.514656686782837,
|
|
"epoch": 2.100418410041841,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004792409060675267,
|
|
"loss": 5.0987,
|
|
"mean_token_accuracy": 0.18958841860294343,
|
|
"num_tokens": 4638729.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.500155210494995,
|
|
"epoch": 2.104602510460251,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004791052829207895,
|
|
"loss": 5.2373,
|
|
"mean_token_accuracy": 0.19260668754577637,
|
|
"num_tokens": 4648759.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.505144500732422,
|
|
"epoch": 2.1087866108786613,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000478969239758195,
|
|
"loss": 5.1915,
|
|
"mean_token_accuracy": 0.19235566407442092,
|
|
"num_tokens": 4657744.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.540197086334229,
|
|
"epoch": 2.1129707112970713,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047883277685969815,
|
|
"loss": 5.146,
|
|
"mean_token_accuracy": 0.1923341080546379,
|
|
"num_tokens": 4666154.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.570902967453003,
|
|
"epoch": 2.1171548117154813,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047869589450611773,
|
|
"loss": 5.1634,
|
|
"mean_token_accuracy": 0.18562294095754622,
|
|
"num_tokens": 4675677.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.45881609916687,
|
|
"epoch": 2.1213389121338913,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004785585929791355,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.18829131871461868,
|
|
"num_tokens": 4684733.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.496757555007934,
|
|
"epoch": 2.1255230125523012,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000478420872561296,
|
|
"loss": 5.08,
|
|
"mean_token_accuracy": 0.1950945407152176,
|
|
"num_tokens": 4694155.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.467521715164184,
|
|
"epoch": 2.1297071129707112,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047828273353600577,
|
|
"loss": 5.102,
|
|
"mean_token_accuracy": 0.19573033601045609,
|
|
"num_tokens": 4703286.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.530854940414429,
|
|
"epoch": 2.1338912133891212,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047814417618753256,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.1914050340652466,
|
|
"num_tokens": 4711666.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.474881649017334,
|
|
"epoch": 2.1380753138075312,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047800520080100526,
|
|
"loss": 5.1598,
|
|
"mean_token_accuracy": 0.18565719872713088,
|
|
"num_tokens": 4721644.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.515645265579224,
|
|
"epoch": 2.1422594142259412,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047786580766241265,
|
|
"loss": 5.1179,
|
|
"mean_token_accuracy": 0.1917526826262474,
|
|
"num_tokens": 4730885.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.460666942596435,
|
|
"epoch": 2.1464435146443517,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004777259970586036,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.19566380977630615,
|
|
"num_tokens": 4740871.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.511660814285278,
|
|
"epoch": 2.1506276150627617,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004775857692772859,
|
|
"loss": 5.0814,
|
|
"mean_token_accuracy": 0.19136700332164763,
|
|
"num_tokens": 4750603.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.451175594329834,
|
|
"epoch": 2.1548117154811717,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004774451246070257,
|
|
"loss": 5.1997,
|
|
"mean_token_accuracy": 0.19287348687648773,
|
|
"num_tokens": 4760681.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.563189554214477,
|
|
"epoch": 2.1589958158995817,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004773040633372471,
|
|
"loss": 5.1697,
|
|
"mean_token_accuracy": 0.19402116388082505,
|
|
"num_tokens": 4770237.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.507778596878052,
|
|
"epoch": 2.1631799163179917,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047716258575823147,
|
|
"loss": 5.1947,
|
|
"mean_token_accuracy": 0.19255647659301758,
|
|
"num_tokens": 4779022.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.568421030044556,
|
|
"epoch": 2.1673640167364017,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000477020692161117,
|
|
"loss": 5.2143,
|
|
"mean_token_accuracy": 0.1894251137971878,
|
|
"num_tokens": 4788644.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.450950956344604,
|
|
"epoch": 2.1715481171548117,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004768783828378978,
|
|
"loss": 5.2274,
|
|
"mean_token_accuracy": 0.1841271325945854,
|
|
"num_tokens": 4798628.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.642622089385986,
|
|
"epoch": 2.1757322175732217,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004767356580814237,
|
|
"loss": 5.3151,
|
|
"mean_token_accuracy": 0.18398940712213516,
|
|
"num_tokens": 4807727.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.572600650787353,
|
|
"epoch": 2.1799163179916317,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004765925181853993,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.1874796748161316,
|
|
"num_tokens": 4817896.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.54551215171814,
|
|
"epoch": 2.1841004184100417,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004764489634443834,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.18952204287052155,
|
|
"num_tokens": 4827135.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.5740166187286375,
|
|
"epoch": 2.1882845188284517,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047630499415378863,
|
|
"loss": 5.2359,
|
|
"mean_token_accuracy": 0.18598488122224807,
|
|
"num_tokens": 4837554.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.446388196945191,
|
|
"epoch": 2.192468619246862,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004761606106098806,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.19046391695737838,
|
|
"num_tokens": 4846739.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.5114672660827635,
|
|
"epoch": 2.196652719665272,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047601581310977747,
|
|
"loss": 5.1708,
|
|
"mean_token_accuracy": 0.19251052439212799,
|
|
"num_tokens": 4855707.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.521695756912232,
|
|
"epoch": 2.200836820083682,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004758706019514493,
|
|
"loss": 5.1731,
|
|
"mean_token_accuracy": 0.19019678086042405,
|
|
"num_tokens": 4865098.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.562310838699341,
|
|
"epoch": 2.205020920502092,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004757249774337172,
|
|
"loss": 5.1816,
|
|
"mean_token_accuracy": 0.19251838326454163,
|
|
"num_tokens": 4874426.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.478124523162842,
|
|
"epoch": 2.209205020920502,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047557893985625293,
|
|
"loss": 5.1378,
|
|
"mean_token_accuracy": 0.19417382776737213,
|
|
"num_tokens": 4883240.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.465423583984375,
|
|
"epoch": 2.213389121338912,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004754324895195786,
|
|
"loss": 5.1202,
|
|
"mean_token_accuracy": 0.19453772753477097,
|
|
"num_tokens": 4891989.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.5380316257476805,
|
|
"epoch": 2.217573221757322,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004752856267250652,
|
|
"loss": 5.2234,
|
|
"mean_token_accuracy": 0.19160993844270707,
|
|
"num_tokens": 4901546.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.401398849487305,
|
|
"epoch": 2.221757322175732,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047513835177493286,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.19670096188783645,
|
|
"num_tokens": 4910222.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.523337650299072,
|
|
"epoch": 2.225941422594142,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004749906649722498,
|
|
"loss": 5.268,
|
|
"mean_token_accuracy": 0.179041288793087,
|
|
"num_tokens": 4920184.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.477303552627563,
|
|
"epoch": 2.2301255230125525,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004748425666209316,
|
|
"loss": 5.1843,
|
|
"mean_token_accuracy": 0.1905567988753319,
|
|
"num_tokens": 4930014.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.487618446350098,
|
|
"epoch": 2.2343096234309625,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004746940570257409,
|
|
"loss": 5.0978,
|
|
"mean_token_accuracy": 0.1955874189734459,
|
|
"num_tokens": 4939556.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.451806926727295,
|
|
"epoch": 2.2384937238493725,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004745451364922867,
|
|
"loss": 5.1513,
|
|
"mean_token_accuracy": 0.18809408247470855,
|
|
"num_tokens": 4949066.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.565406703948975,
|
|
"epoch": 2.2426778242677825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004743958053270233,
|
|
"loss": 5.1631,
|
|
"mean_token_accuracy": 0.19298059940338136,
|
|
"num_tokens": 4957881.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.498418807983398,
|
|
"epoch": 2.2468619246861925,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047424606383725037,
|
|
"loss": 5.2088,
|
|
"mean_token_accuracy": 0.1867537796497345,
|
|
"num_tokens": 4967105.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.4907450675964355,
|
|
"epoch": 2.2510460251046025,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004740959123311118,
|
|
"loss": 5.1032,
|
|
"mean_token_accuracy": 0.1962350770831108,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.47246789932251,
|
|
"epoch": 2.2552301255230125,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047394535111759526,
|
|
"loss": 5.1332,
|
|
"mean_token_accuracy": 0.19636671543121337,
|
|
"num_tokens": 4986163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.462388801574707,
|
|
"epoch": 2.2594142259414225,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004737943805065314,
|
|
"loss": 5.1283,
|
|
"mean_token_accuracy": 0.19069456607103347,
|
|
"num_tokens": 4995179.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.5717652320861815,
|
|
"epoch": 2.2635983263598325,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047364300080859363,
|
|
"loss": 5.2184,
|
|
"mean_token_accuracy": 0.1857464611530304,
|
|
"num_tokens": 5003856.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.496053695678711,
|
|
"epoch": 2.2677824267782425,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047349121233529695,
|
|
"loss": 5.2509,
|
|
"mean_token_accuracy": 0.18558135330677034,
|
|
"num_tokens": 5013506.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.482390022277832,
|
|
"epoch": 2.2719665271966525,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004733390153989976,
|
|
"loss": 5.183,
|
|
"mean_token_accuracy": 0.19485190361738206,
|
|
"num_tokens": 5022159.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.535829067230225,
|
|
"epoch": 2.276150627615063,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047318641031289245,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.19356691241264343,
|
|
"num_tokens": 5031869.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.3320960521698,
|
|
"epoch": 2.280334728033473,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004730333973910183,
|
|
"loss": 4.9986,
|
|
"mean_token_accuracy": 0.2023811861872673,
|
|
"num_tokens": 5039904.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.533828926086426,
|
|
"epoch": 2.284518828451883,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004728799769482511,
|
|
"loss": 5.2671,
|
|
"mean_token_accuracy": 0.18972275108098985,
|
|
"num_tokens": 5049144.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.51212067604065,
|
|
"epoch": 2.288702928870293,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047272614930030533,
|
|
"loss": 5.0725,
|
|
"mean_token_accuracy": 0.19585145264863968,
|
|
"num_tokens": 5058302.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.443295001983643,
|
|
"epoch": 2.292887029288703,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047257191476373375,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.1954988420009613,
|
|
"num_tokens": 5067443.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.445563745498657,
|
|
"epoch": 2.297071129707113,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047241727365592603,
|
|
"loss": 5.2138,
|
|
"mean_token_accuracy": 0.18225188106298446,
|
|
"num_tokens": 5076415.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.524112510681152,
|
|
"epoch": 2.301255230125523,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004722622262951089,
|
|
"loss": 5.1806,
|
|
"mean_token_accuracy": 0.1864774376153946,
|
|
"num_tokens": 5085882.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.525083065032959,
|
|
"epoch": 2.305439330543933,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004721067730003448,
|
|
"loss": 5.1394,
|
|
"mean_token_accuracy": 0.20481592714786528,
|
|
"num_tokens": 5094685.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.449747610092163,
|
|
"epoch": 2.309623430962343,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004719509140915317,
|
|
"loss": 5.1596,
|
|
"mean_token_accuracy": 0.1895164042711258,
|
|
"num_tokens": 5103885.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.492194175720215,
|
|
"epoch": 2.3138075313807533,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047179464988940205,
|
|
"loss": 5.0871,
|
|
"mean_token_accuracy": 0.19752362817525865,
|
|
"num_tokens": 5112164.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.43505220413208,
|
|
"epoch": 2.3179916317991633,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047163798071552253,
|
|
"loss": 5.0694,
|
|
"mean_token_accuracy": 0.1989555761218071,
|
|
"num_tokens": 5121913.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.412868309020996,
|
|
"epoch": 2.3221757322175733,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004714809068922931,
|
|
"loss": 5.0596,
|
|
"mean_token_accuracy": 0.19883574396371842,
|
|
"num_tokens": 5130447.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.469843626022339,
|
|
"epoch": 2.3263598326359833,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004713234287429465,
|
|
"loss": 5.1999,
|
|
"mean_token_accuracy": 0.18897882103919983,
|
|
"num_tokens": 5139863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.4477434158325195,
|
|
"epoch": 2.3305439330543933,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004711655465915473,
|
|
"loss": 5.1573,
|
|
"mean_token_accuracy": 0.19469541758298875,
|
|
"num_tokens": 5149945.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.517200565338134,
|
|
"epoch": 2.3347280334728033,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004710072607629917,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.18986839056015015,
|
|
"num_tokens": 5160293.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.376310396194458,
|
|
"epoch": 2.3389121338912133,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047084857158300646,
|
|
"loss": 5.1176,
|
|
"mean_token_accuracy": 0.18341436684131623,
|
|
"num_tokens": 5169896.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.495307111740113,
|
|
"epoch": 2.3430962343096233,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047068947937814844,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.18457840234041215,
|
|
"num_tokens": 5178898.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.514347553253174,
|
|
"epoch": 2.3472803347280333,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047052998447580374,
|
|
"loss": 5.2235,
|
|
"mean_token_accuracy": 0.18590953797101975,
|
|
"num_tokens": 5188733.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.433951377868652,
|
|
"epoch": 2.3514644351464433,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047037008720418724,
|
|
"loss": 5.1382,
|
|
"mean_token_accuracy": 0.19282420724630356,
|
|
"num_tokens": 5198117.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.539768171310425,
|
|
"epoch": 2.3556485355648533,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004702097878923419,
|
|
"loss": 5.1761,
|
|
"mean_token_accuracy": 0.18806317895650865,
|
|
"num_tokens": 5206791.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.4410144805908205,
|
|
"epoch": 2.3598326359832638,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004700490868701377,
|
|
"loss": 5.0661,
|
|
"mean_token_accuracy": 0.19899241179227828,
|
|
"num_tokens": 5216427.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.3441893577575685,
|
|
"epoch": 2.3640167364016738,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004698879844682718,
|
|
"loss": 5.0688,
|
|
"mean_token_accuracy": 0.19825328886508942,
|
|
"num_tokens": 5225983.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.419188642501831,
|
|
"epoch": 2.3682008368200838,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046972648101826695,
|
|
"loss": 5.1377,
|
|
"mean_token_accuracy": 0.19336697161197663,
|
|
"num_tokens": 5234547.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.562987756729126,
|
|
"epoch": 2.3723849372384938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004695645768524712,
|
|
"loss": 5.1905,
|
|
"mean_token_accuracy": 0.19764263331890106,
|
|
"num_tokens": 5243281.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.465346097946167,
|
|
"epoch": 2.3765690376569037,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004694022723040573,
|
|
"loss": 5.1063,
|
|
"mean_token_accuracy": 0.18801086246967316,
|
|
"num_tokens": 5252318.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.477588272094726,
|
|
"epoch": 2.3807531380753137,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046923956770702206,
|
|
"loss": 5.1812,
|
|
"mean_token_accuracy": 0.19431206732988357,
|
|
"num_tokens": 5261781.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.525727462768555,
|
|
"epoch": 2.3849372384937237,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004690764633961852,
|
|
"loss": 5.0792,
|
|
"mean_token_accuracy": 0.1964922234416008,
|
|
"num_tokens": 5270409.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.320622968673706,
|
|
"epoch": 2.3891213389121337,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004689129597071893,
|
|
"loss": 5.0885,
|
|
"mean_token_accuracy": 0.20231516063213348,
|
|
"num_tokens": 5278766.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.494616937637329,
|
|
"epoch": 2.393305439330544,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046874905697649857,
|
|
"loss": 5.2489,
|
|
"mean_token_accuracy": 0.1810278281569481,
|
|
"num_tokens": 5287572.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.48681230545044,
|
|
"epoch": 2.397489539748954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046858475554139857,
|
|
"loss": 5.1476,
|
|
"mean_token_accuracy": 0.19416207671165467,
|
|
"num_tokens": 5297348.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.5345282554626465,
|
|
"epoch": 2.401673640167364,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004684200557399952,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.19453701078891755,
|
|
"num_tokens": 5305748.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.427400588989258,
|
|
"epoch": 2.405857740585774,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046825495791121414,
|
|
"loss": 5.1168,
|
|
"mean_token_accuracy": 0.19719472527503967,
|
|
"num_tokens": 5314482.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.463809633255005,
|
|
"epoch": 2.410041841004184,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046808946239480017,
|
|
"loss": 5.212,
|
|
"mean_token_accuracy": 0.18592767417430878,
|
|
"num_tokens": 5325161.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.398594665527344,
|
|
"epoch": 2.414225941422594,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004679235695313165,
|
|
"loss": 5.19,
|
|
"mean_token_accuracy": 0.18461235761642455,
|
|
"num_tokens": 5335262.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.4714151382446286,
|
|
"epoch": 2.418410041841004,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046775727966214393,
|
|
"loss": 5.0901,
|
|
"mean_token_accuracy": 0.18747085183858872,
|
|
"num_tokens": 5345145.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.37877402305603,
|
|
"epoch": 2.422594142259414,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004675905931294803,
|
|
"loss": 5.0786,
|
|
"mean_token_accuracy": 0.19339294284582137,
|
|
"num_tokens": 5354715.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.385033226013183,
|
|
"epoch": 2.426778242677824,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004674235102763398,
|
|
"loss": 5.1354,
|
|
"mean_token_accuracy": 0.19722559601068496,
|
|
"num_tokens": 5364180.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.409422111511231,
|
|
"epoch": 2.430962343096234,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046725603144655195,
|
|
"loss": 5.1049,
|
|
"mean_token_accuracy": 0.1981211021542549,
|
|
"num_tokens": 5372919.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.408568286895752,
|
|
"epoch": 2.435146443514644,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004670881569847612,
|
|
"loss": 5.081,
|
|
"mean_token_accuracy": 0.19983326643705368,
|
|
"num_tokens": 5383002.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.561278867721557,
|
|
"epoch": 2.439330543933054,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046691988723642636,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.18554889261722565,
|
|
"num_tokens": 5393622.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.4837995052337645,
|
|
"epoch": 2.4435146443514646,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004667512225478195,
|
|
"loss": 5.1679,
|
|
"mean_token_accuracy": 0.19134593456983567,
|
|
"num_tokens": 5403106.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.487650346755982,
|
|
"epoch": 2.4476987447698746,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046658216326602526,
|
|
"loss": 5.1458,
|
|
"mean_token_accuracy": 0.1922094628214836,
|
|
"num_tokens": 5412596.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.488178873062134,
|
|
"epoch": 2.4518828451882846,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004664127097389406,
|
|
"loss": 5.1866,
|
|
"mean_token_accuracy": 0.19006673842668534,
|
|
"num_tokens": 5422301.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.522394418716431,
|
|
"epoch": 2.4560669456066946,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046624286231527367,
|
|
"loss": 5.1078,
|
|
"mean_token_accuracy": 0.19487448036670685,
|
|
"num_tokens": 5432178.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.365673065185547,
|
|
"epoch": 2.4602510460251046,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004660726213445431,
|
|
"loss": 5.1563,
|
|
"mean_token_accuracy": 0.19536193758249282,
|
|
"num_tokens": 5441162.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.394482612609863,
|
|
"epoch": 2.4644351464435146,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004659019871770774,
|
|
"loss": 5.0707,
|
|
"mean_token_accuracy": 0.2075057402253151,
|
|
"num_tokens": 5450566.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.459809160232544,
|
|
"epoch": 2.4686192468619246,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046573096016401435,
|
|
"loss": 5.1164,
|
|
"mean_token_accuracy": 0.2048611044883728,
|
|
"num_tokens": 5460289.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.367074680328369,
|
|
"epoch": 2.4728033472803346,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004655595406572999,
|
|
"loss": 5.0942,
|
|
"mean_token_accuracy": 0.19705163538455964,
|
|
"num_tokens": 5469581.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.467730760574341,
|
|
"epoch": 2.476987447698745,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000465387729009688,
|
|
"loss": 5.1978,
|
|
"mean_token_accuracy": 0.195025996863842,
|
|
"num_tokens": 5478005.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.444231843948364,
|
|
"epoch": 2.481171548117155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046521552557473927,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.1942666307091713,
|
|
"num_tokens": 5486963.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.377353382110596,
|
|
"epoch": 2.485355648535565,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046504293070682073,
|
|
"loss": 5.0599,
|
|
"mean_token_accuracy": 0.19242556393146515,
|
|
"num_tokens": 5496368.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.39252119064331,
|
|
"epoch": 2.489539748953975,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046486994476110493,
|
|
"loss": 5.0683,
|
|
"mean_token_accuracy": 0.19476308226585387,
|
|
"num_tokens": 5505706.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.359625625610351,
|
|
"epoch": 2.493723849372385,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046469656809356916,
|
|
"loss": 5.0492,
|
|
"mean_token_accuracy": 0.2019849270582199,
|
|
"num_tokens": 5513945.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.372553396224975,
|
|
"epoch": 2.497907949790795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046452280106099454,
|
|
"loss": 5.0995,
|
|
"mean_token_accuracy": 0.19284075796604155,
|
|
"num_tokens": 5523831.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.487049913406372,
|
|
"epoch": 2.502092050209205,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004643486440209659,
|
|
"loss": 5.2515,
|
|
"mean_token_accuracy": 0.1912768691778183,
|
|
"num_tokens": 5532705.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.466883277893066,
|
|
"epoch": 2.506276150627615,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004641740973318702,
|
|
"loss": 5.1617,
|
|
"mean_token_accuracy": 0.19773912131786348,
|
|
"num_tokens": 5542793.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.490420913696289,
|
|
"epoch": 2.510460251046025,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004639991613528967,
|
|
"loss": 5.1586,
|
|
"mean_token_accuracy": 0.1931561604142189,
|
|
"num_tokens": 5551938.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.510460251046025,
|
|
"eval_entropy": 5.16201673304864,
|
|
"eval_loss": 5.4512505531311035,
|
|
"eval_mean_token_accuracy": 0.18672356001865156,
|
|
"eval_num_tokens": 5551938.0,
|
|
"eval_runtime": 2.0984,
|
|
"eval_samples_per_second": 1861.931,
|
|
"eval_steps_per_second": 233.039,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.425399446487427,
|
|
"epoch": 2.514644351464435,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046382383644403543,
|
|
"loss": 5.1835,
|
|
"mean_token_accuracy": 0.19039839506149292,
|
|
"num_tokens": 5561374.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.339206981658935,
|
|
"epoch": 2.518828451882845,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046364812296607676,
|
|
"loss": 5.0501,
|
|
"mean_token_accuracy": 0.1937744453549385,
|
|
"num_tokens": 5570459.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.338240003585815,
|
|
"epoch": 2.523012552301255,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046347202128061094,
|
|
"loss": 5.1458,
|
|
"mean_token_accuracy": 0.19012720435857772,
|
|
"num_tokens": 5579692.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.523472547531128,
|
|
"epoch": 2.5271966527196654,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004632955317500268,
|
|
"loss": 5.1118,
|
|
"mean_token_accuracy": 0.1933310493826866,
|
|
"num_tokens": 5588338.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.4029021739959715,
|
|
"epoch": 2.5313807531380754,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046311865473751154,
|
|
"loss": 5.1825,
|
|
"mean_token_accuracy": 0.19432860165834426,
|
|
"num_tokens": 5598589.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.435482406616211,
|
|
"epoch": 2.5355648535564854,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004629413906070495,
|
|
"loss": 5.1497,
|
|
"mean_token_accuracy": 0.20009873807430267,
|
|
"num_tokens": 5608127.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.502504062652588,
|
|
"epoch": 2.5397489539748954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046276373972342177,
|
|
"loss": 5.217,
|
|
"mean_token_accuracy": 0.19778114706277847,
|
|
"num_tokens": 5617728.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.3748516082763675,
|
|
"epoch": 2.5439330543933054,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004625857024522054,
|
|
"loss": 5.1472,
|
|
"mean_token_accuracy": 0.19480963200330734,
|
|
"num_tokens": 5626823.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.403126430511475,
|
|
"epoch": 2.5481171548117154,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046240727915977236,
|
|
"loss": 5.1335,
|
|
"mean_token_accuracy": 0.19545071423053742,
|
|
"num_tokens": 5636284.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.50823016166687,
|
|
"epoch": 2.5523012552301254,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004622284702132893,
|
|
"loss": 5.2651,
|
|
"mean_token_accuracy": 0.18839738219976426,
|
|
"num_tokens": 5645721.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.421586608886718,
|
|
"epoch": 2.5564853556485354,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046204927598071595,
|
|
"loss": 5.1708,
|
|
"mean_token_accuracy": 0.19611793607473374,
|
|
"num_tokens": 5654734.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.4399957180023195,
|
|
"epoch": 2.560669456066946,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046186969683080547,
|
|
"loss": 5.1607,
|
|
"mean_token_accuracy": 0.20011166483163834,
|
|
"num_tokens": 5664010.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.42301082611084,
|
|
"epoch": 2.564853556485356,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046168973313310294,
|
|
"loss": 5.1476,
|
|
"mean_token_accuracy": 0.1952656924724579,
|
|
"num_tokens": 5673028.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.425078392028809,
|
|
"epoch": 2.569037656903766,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004615093852579445,
|
|
"loss": 5.1075,
|
|
"mean_token_accuracy": 0.19344520270824433,
|
|
"num_tokens": 5682517.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.317914009094238,
|
|
"epoch": 2.573221757322176,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046132865357645717,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.20334567576646806,
|
|
"num_tokens": 5690920.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.397665357589721,
|
|
"epoch": 2.577405857740586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046114753846055774,
|
|
"loss": 5.1139,
|
|
"mean_token_accuracy": 0.19354642778635026,
|
|
"num_tokens": 5700336.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.473945760726929,
|
|
"epoch": 2.581589958158996,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000460966040282952,
|
|
"loss": 5.2596,
|
|
"mean_token_accuracy": 0.1867171883583069,
|
|
"num_tokens": 5711085.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.438890409469605,
|
|
"epoch": 2.585774058577406,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046078415941713384,
|
|
"loss": 5.0669,
|
|
"mean_token_accuracy": 0.20374951660633087,
|
|
"num_tokens": 5719560.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.340834712982177,
|
|
"epoch": 2.589958158995816,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004606018962373852,
|
|
"loss": 5.0312,
|
|
"mean_token_accuracy": 0.20423560887575148,
|
|
"num_tokens": 5727865.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.388959884643555,
|
|
"epoch": 2.594142259414226,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004604192511187741,
|
|
"loss": 5.1505,
|
|
"mean_token_accuracy": 0.18716893047094346,
|
|
"num_tokens": 5737705.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.469685649871826,
|
|
"epoch": 2.598326359832636,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046023622443715483,
|
|
"loss": 5.1842,
|
|
"mean_token_accuracy": 0.18010880798101425,
|
|
"num_tokens": 5747057.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.391718959808349,
|
|
"epoch": 2.602510460251046,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046005281656916707,
|
|
"loss": 5.1385,
|
|
"mean_token_accuracy": 0.1927603155374527,
|
|
"num_tokens": 5755793.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.420816326141358,
|
|
"epoch": 2.606694560669456,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004598690278922346,
|
|
"loss": 5.1212,
|
|
"mean_token_accuracy": 0.1923212856054306,
|
|
"num_tokens": 5765385.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.3141584396362305,
|
|
"epoch": 2.6108786610878663,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045968485878456497,
|
|
"loss": 5.0621,
|
|
"mean_token_accuracy": 0.19610344469547272,
|
|
"num_tokens": 5773947.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.432509851455689,
|
|
"epoch": 2.6150627615062763,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004595003096251486,
|
|
"loss": 5.0486,
|
|
"mean_token_accuracy": 0.19920838326215745,
|
|
"num_tokens": 5782700.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.480187320709229,
|
|
"epoch": 2.6192468619246863,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045931538079375803,
|
|
"loss": 5.2179,
|
|
"mean_token_accuracy": 0.18359574377536775,
|
|
"num_tokens": 5792452.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.437179660797119,
|
|
"epoch": 2.6234309623430963,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045913007267094704,
|
|
"loss": 5.1515,
|
|
"mean_token_accuracy": 0.19422263354063035,
|
|
"num_tokens": 5801270.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.473179864883423,
|
|
"epoch": 2.6276150627615062,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004589443856380499,
|
|
"loss": 5.176,
|
|
"mean_token_accuracy": 0.2000298708677292,
|
|
"num_tokens": 5810704.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.434019231796265,
|
|
"epoch": 2.6317991631799162,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004587583200771807,
|
|
"loss": 5.1326,
|
|
"mean_token_accuracy": 0.1936563014984131,
|
|
"num_tokens": 5819690.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.357477807998658,
|
|
"epoch": 2.6359832635983262,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004585718763712325,
|
|
"loss": 5.0521,
|
|
"mean_token_accuracy": 0.1926045000553131,
|
|
"num_tokens": 5828912.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.275949287414551,
|
|
"epoch": 2.6401673640167362,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004583850549038764,
|
|
"loss": 4.995,
|
|
"mean_token_accuracy": 0.19906808137893678,
|
|
"num_tokens": 5836914.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.407176065444946,
|
|
"epoch": 2.6443514644351467,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045819785605956105,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.19743258953094484,
|
|
"num_tokens": 5846268.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.408795166015625,
|
|
"epoch": 2.6485355648535567,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004580102802235115,
|
|
"loss": 5.1517,
|
|
"mean_token_accuracy": 0.1956247165799141,
|
|
"num_tokens": 5855029.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.440755558013916,
|
|
"epoch": 2.6527196652719667,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004578223277817286,
|
|
"loss": 5.1442,
|
|
"mean_token_accuracy": 0.19061772674322128,
|
|
"num_tokens": 5865049.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.505774068832397,
|
|
"epoch": 2.6569037656903767,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045763399912098835,
|
|
"loss": 5.1451,
|
|
"mean_token_accuracy": 0.19250471740961075,
|
|
"num_tokens": 5873593.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.33617901802063,
|
|
"epoch": 2.6610878661087867,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045744529462884086,
|
|
"loss": 5.0583,
|
|
"mean_token_accuracy": 0.19942415803670882,
|
|
"num_tokens": 5882562.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.376270532608032,
|
|
"epoch": 2.6652719665271967,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045725621469360964,
|
|
"loss": 5.2511,
|
|
"mean_token_accuracy": 0.1823133036494255,
|
|
"num_tokens": 5892516.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.460144138336181,
|
|
"epoch": 2.6694560669456067,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004570667597043907,
|
|
"loss": 5.1331,
|
|
"mean_token_accuracy": 0.19214058369398118,
|
|
"num_tokens": 5901494.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.41216778755188,
|
|
"epoch": 2.6736401673640167,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000456876930051052,
|
|
"loss": 5.121,
|
|
"mean_token_accuracy": 0.19413231313228607,
|
|
"num_tokens": 5910750.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.491813993453979,
|
|
"epoch": 2.6778242677824267,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004566867261242325,
|
|
"loss": 5.1513,
|
|
"mean_token_accuracy": 0.19640996754169465,
|
|
"num_tokens": 5920277.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.376474189758301,
|
|
"epoch": 2.6820083682008367,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045649614831534127,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.19329559653997422,
|
|
"num_tokens": 5929880.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.510521030426025,
|
|
"epoch": 2.6861924686192467,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045630519701655677,
|
|
"loss": 5.2887,
|
|
"mean_token_accuracy": 0.1889829695224762,
|
|
"num_tokens": 5939538.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.437780284881592,
|
|
"epoch": 2.6903765690376567,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045611387262082604,
|
|
"loss": 5.1221,
|
|
"mean_token_accuracy": 0.19535306692123414,
|
|
"num_tokens": 5948747.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.377941036224366,
|
|
"epoch": 2.694560669456067,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045592217552186407,
|
|
"loss": 5.1128,
|
|
"mean_token_accuracy": 0.1961102232336998,
|
|
"num_tokens": 5958209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.458626556396484,
|
|
"epoch": 2.698744769874477,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004557301061141525,
|
|
"loss": 5.1538,
|
|
"mean_token_accuracy": 0.19718127101659774,
|
|
"num_tokens": 5967184.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.343305158615112,
|
|
"epoch": 2.702928870292887,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045553766479293945,
|
|
"loss": 5.1693,
|
|
"mean_token_accuracy": 0.18589765280485154,
|
|
"num_tokens": 5976006.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.304282236099243,
|
|
"epoch": 2.707112970711297,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045534485195423806,
|
|
"loss": 5.0101,
|
|
"mean_token_accuracy": 0.2017744779586792,
|
|
"num_tokens": 5984671.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.51267147064209,
|
|
"epoch": 2.711297071129707,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045515166799482633,
|
|
"loss": 5.1474,
|
|
"mean_token_accuracy": 0.1944565862417221,
|
|
"num_tokens": 5993190.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.413865232467652,
|
|
"epoch": 2.715481171548117,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045495811331224564,
|
|
"loss": 5.1434,
|
|
"mean_token_accuracy": 0.19161706119775773,
|
|
"num_tokens": 6004080.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.400912714004517,
|
|
"epoch": 2.719665271966527,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004547641883048005,
|
|
"loss": 5.1282,
|
|
"mean_token_accuracy": 0.19718449264764787,
|
|
"num_tokens": 6013212.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.399621391296387,
|
|
"epoch": 2.723849372384937,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004545698933715573,
|
|
"loss": 5.0664,
|
|
"mean_token_accuracy": 0.20179840326309204,
|
|
"num_tokens": 6023444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.449490594863891,
|
|
"epoch": 2.7280334728033475,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045437522891234386,
|
|
"loss": 5.0807,
|
|
"mean_token_accuracy": 0.19586815387010575,
|
|
"num_tokens": 6032217.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.377200269699097,
|
|
"epoch": 2.7322175732217575,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045418019532774827,
|
|
"loss": 5.0333,
|
|
"mean_token_accuracy": 0.20085300803184508,
|
|
"num_tokens": 6040246.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.38801851272583,
|
|
"epoch": 2.7364016736401675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045398479301911836,
|
|
"loss": 5.1536,
|
|
"mean_token_accuracy": 0.18894162625074387,
|
|
"num_tokens": 6049447.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.288433313369751,
|
|
"epoch": 2.7405857740585775,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045378902238856064,
|
|
"loss": 5.029,
|
|
"mean_token_accuracy": 0.20027020275592805,
|
|
"num_tokens": 6058294.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.398590087890625,
|
|
"epoch": 2.7447698744769875,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004535928838389396,
|
|
"loss": 5.1699,
|
|
"mean_token_accuracy": 0.19035880416631698,
|
|
"num_tokens": 6067237.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.497797870635987,
|
|
"epoch": 2.7489539748953975,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004533963777738768,
|
|
"loss": 5.1607,
|
|
"mean_token_accuracy": 0.19322109073400498,
|
|
"num_tokens": 6075599.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.44541711807251,
|
|
"epoch": 2.7531380753138075,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045319950459775024,
|
|
"loss": 5.1382,
|
|
"mean_token_accuracy": 0.18743716329336166,
|
|
"num_tokens": 6085131.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.4336793422698975,
|
|
"epoch": 2.7573221757322175,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004530022647156931,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.2008256882429123,
|
|
"num_tokens": 6094745.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.4262611865997314,
|
|
"epoch": 2.7615062761506275,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004528046585335935,
|
|
"loss": 5.1772,
|
|
"mean_token_accuracy": 0.18417732119560243,
|
|
"num_tokens": 6104182.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.388008737564087,
|
|
"epoch": 2.7656903765690375,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045260668645809304,
|
|
"loss": 5.028,
|
|
"mean_token_accuracy": 0.20391566306352615,
|
|
"num_tokens": 6113370.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.338034915924072,
|
|
"epoch": 2.7698744769874475,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004524083488965866,
|
|
"loss": 5.1122,
|
|
"mean_token_accuracy": 0.1975890129804611,
|
|
"num_tokens": 6122202.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.3651535987854,
|
|
"epoch": 2.7740585774058575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004522096462572208,
|
|
"loss": 5.154,
|
|
"mean_token_accuracy": 0.19045838564634324,
|
|
"num_tokens": 6131081.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.549214553833008,
|
|
"epoch": 2.778242677824268,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004520105789488939,
|
|
"loss": 5.2344,
|
|
"mean_token_accuracy": 0.18843050450086593,
|
|
"num_tokens": 6141707.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.376821041107178,
|
|
"epoch": 2.782426778242678,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045181114738125427,
|
|
"loss": 5.0798,
|
|
"mean_token_accuracy": 0.1986973837018013,
|
|
"num_tokens": 6151651.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.390552663803101,
|
|
"epoch": 2.786610878661088,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045161135196470015,
|
|
"loss": 5.0459,
|
|
"mean_token_accuracy": 0.20318312346935272,
|
|
"num_tokens": 6159928.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.370474720001221,
|
|
"epoch": 2.790794979079498,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004514111931103783,
|
|
"loss": 5.1461,
|
|
"mean_token_accuracy": 0.19062791317701339,
|
|
"num_tokens": 6169115.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.4296621799469,
|
|
"epoch": 2.794979079497908,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045121067123018353,
|
|
"loss": 5.1614,
|
|
"mean_token_accuracy": 0.19591851383447648,
|
|
"num_tokens": 6177995.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.444316434860229,
|
|
"epoch": 2.799163179916318,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045100978673675755,
|
|
"loss": 5.1483,
|
|
"mean_token_accuracy": 0.19177227467298508,
|
|
"num_tokens": 6188081.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.452777671813965,
|
|
"epoch": 2.803347280334728,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045080854004348836,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.19564290046691896,
|
|
"num_tokens": 6198010.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.315631151199341,
|
|
"epoch": 2.8075313807531384,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045060693156450933,
|
|
"loss": 5.0976,
|
|
"mean_token_accuracy": 0.19865680038928984,
|
|
"num_tokens": 6207164.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.441837358474731,
|
|
"epoch": 2.8117154811715483,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004504049617146983,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.1870425045490265,
|
|
"num_tokens": 6216512.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.403610372543335,
|
|
"epoch": 2.8158995815899583,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045020263090967674,
|
|
"loss": 5.1691,
|
|
"mean_token_accuracy": 0.19616774767637252,
|
|
"num_tokens": 6226196.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.418008041381836,
|
|
"epoch": 2.8200836820083683,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004499999395658088,
|
|
"loss": 5.043,
|
|
"mean_token_accuracy": 0.1955062434077263,
|
|
"num_tokens": 6234820.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.327598428726196,
|
|
"epoch": 2.8242677824267783,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000449796888100201,
|
|
"loss": 5.1221,
|
|
"mean_token_accuracy": 0.19041236042976378,
|
|
"num_tokens": 6243813.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.5224442958831785,
|
|
"epoch": 2.8284518828451883,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044959347693070026,
|
|
"loss": 5.2074,
|
|
"mean_token_accuracy": 0.18874130994081498,
|
|
"num_tokens": 6253376.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.506325721740723,
|
|
"epoch": 2.8326359832635983,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004493897064758943,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.19199124872684478,
|
|
"num_tokens": 6262640.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.435546255111694,
|
|
"epoch": 2.8368200836820083,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004491855771551098,
|
|
"loss": 5.1722,
|
|
"mean_token_accuracy": 0.19521231800317765,
|
|
"num_tokens": 6270908.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.374811601638794,
|
|
"epoch": 2.8410041841004183,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004489810893884121,
|
|
"loss": 5.0847,
|
|
"mean_token_accuracy": 0.19597941190004348,
|
|
"num_tokens": 6280361.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.3854241371154785,
|
|
"epoch": 2.8451882845188283,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004487762435966042,
|
|
"loss": 5.1003,
|
|
"mean_token_accuracy": 0.1955694168806076,
|
|
"num_tokens": 6288832.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.342634582519532,
|
|
"epoch": 2.8493723849372383,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044857104020122585,
|
|
"loss": 5.0837,
|
|
"mean_token_accuracy": 0.19656772315502166,
|
|
"num_tokens": 6298553.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.328307151794434,
|
|
"epoch": 2.8535564853556483,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004483654796245526,
|
|
"loss": 5.0272,
|
|
"mean_token_accuracy": 0.2013445347547531,
|
|
"num_tokens": 6307692.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.431701564788819,
|
|
"epoch": 2.8577405857740583,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000448159562289595,
|
|
"loss": 5.1173,
|
|
"mean_token_accuracy": 0.19215036183595657,
|
|
"num_tokens": 6316729.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.45587568283081,
|
|
"epoch": 2.8619246861924688,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044795328862009794,
|
|
"loss": 5.1991,
|
|
"mean_token_accuracy": 0.18768998235464096,
|
|
"num_tokens": 6326210.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.288549852371216,
|
|
"epoch": 2.8661087866108788,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044774665904053943,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.19746658951044083,
|
|
"num_tokens": 6335156.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.3288407802581785,
|
|
"epoch": 2.8702928870292888,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004475396739761299,
|
|
"loss": 4.9992,
|
|
"mean_token_accuracy": 0.20105221420526503,
|
|
"num_tokens": 6344264.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.386258697509765,
|
|
"epoch": 2.8744769874476988,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044733233385281117,
|
|
"loss": 5.0598,
|
|
"mean_token_accuracy": 0.20211081802845002,
|
|
"num_tokens": 6352997.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.330745029449463,
|
|
"epoch": 2.8786610878661087,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044712463909725626,
|
|
"loss": 5.0753,
|
|
"mean_token_accuracy": 0.1971671387553215,
|
|
"num_tokens": 6362312.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.493720197677613,
|
|
"epoch": 2.8828451882845187,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044691659013686727,
|
|
"loss": 5.1902,
|
|
"mean_token_accuracy": 0.18817922919988633,
|
|
"num_tokens": 6372107.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.368692445755005,
|
|
"epoch": 2.8870292887029287,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004467081873997756,
|
|
"loss": 5.0324,
|
|
"mean_token_accuracy": 0.2046805202960968,
|
|
"num_tokens": 6381306.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.317897462844849,
|
|
"epoch": 2.891213389121339,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044649943131484056,
|
|
"loss": 4.9818,
|
|
"mean_token_accuracy": 0.1953648269176483,
|
|
"num_tokens": 6390404.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.46919322013855,
|
|
"epoch": 2.895397489539749,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044629032231164874,
|
|
"loss": 5.1934,
|
|
"mean_token_accuracy": 0.1884245216846466,
|
|
"num_tokens": 6400366.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.395158767700195,
|
|
"epoch": 2.899581589958159,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004460808608205127,
|
|
"loss": 5.011,
|
|
"mean_token_accuracy": 0.20137243270874022,
|
|
"num_tokens": 6408955.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.275506830215454,
|
|
"epoch": 2.903765690376569,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004458710472724706,
|
|
"loss": 5.0549,
|
|
"mean_token_accuracy": 0.1969423845410347,
|
|
"num_tokens": 6417172.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.375492334365845,
|
|
"epoch": 2.907949790794979,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044566088209928485,
|
|
"loss": 5.0759,
|
|
"mean_token_accuracy": 0.19510919004678726,
|
|
"num_tokens": 6425898.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.349626636505127,
|
|
"epoch": 2.912133891213389,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00044545036573344175,
|
|
"loss": 5.121,
|
|
"mean_token_accuracy": 0.19467864632606507,
|
|
"num_tokens": 6435940.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.339689493179321,
|
|
"epoch": 2.916317991631799,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044523949860815015,
|
|
"loss": 5.0427,
|
|
"mean_token_accuracy": 0.20257246047258376,
|
|
"num_tokens": 6445119.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.343971252441406,
|
|
"epoch": 2.920502092050209,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004450282811573405,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.19712676256895065,
|
|
"num_tokens": 6453986.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.371427345275879,
|
|
"epoch": 2.924686192468619,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044481671381566446,
|
|
"loss": 5.061,
|
|
"mean_token_accuracy": 0.19536331593990325,
|
|
"num_tokens": 6463658.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.371505212783814,
|
|
"epoch": 2.928870292887029,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004446047970184937,
|
|
"loss": 5.1085,
|
|
"mean_token_accuracy": 0.19880263954401017,
|
|
"num_tokens": 6472370.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.928870292887029,
|
|
"eval_entropy": 5.126956300013879,
|
|
"eval_loss": 5.377480983734131,
|
|
"eval_mean_token_accuracy": 0.19094371527494342,
|
|
"eval_num_tokens": 6472370.0,
|
|
"eval_runtime": 2.0836,
|
|
"eval_samples_per_second": 1875.081,
|
|
"eval_steps_per_second": 234.685,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.422401475906372,
|
|
"epoch": 2.933054393305439,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044439253120191877,
|
|
"loss": 5.2291,
|
|
"mean_token_accuracy": 0.1850748360157013,
|
|
"num_tokens": 6482642.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.391531324386596,
|
|
"epoch": 2.937238493723849,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044417991680274874,
|
|
"loss": 5.1266,
|
|
"mean_token_accuracy": 0.19513725787401198,
|
|
"num_tokens": 6491818.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.374328708648681,
|
|
"epoch": 2.941422594142259,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004439669542585097,
|
|
"loss": 5.112,
|
|
"mean_token_accuracy": 0.1928705170750618,
|
|
"num_tokens": 6500908.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.347389268875122,
|
|
"epoch": 2.9456066945606696,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004437536440074444,
|
|
"loss": 5.0861,
|
|
"mean_token_accuracy": 0.19696730077266694,
|
|
"num_tokens": 6510182.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.335335636138916,
|
|
"epoch": 2.9497907949790796,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004435399864885111,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.1988907501101494,
|
|
"num_tokens": 6519051.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.283881139755249,
|
|
"epoch": 2.9539748953974896,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004433259821413825,
|
|
"loss": 5.0046,
|
|
"mean_token_accuracy": 0.20833035856485366,
|
|
"num_tokens": 6527949.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.50003457069397,
|
|
"epoch": 2.9581589958158996,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044311163140644524,
|
|
"loss": 5.2705,
|
|
"mean_token_accuracy": 0.1907091408967972,
|
|
"num_tokens": 6537227.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.370486927032471,
|
|
"epoch": 2.9623430962343096,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044289693472479867,
|
|
"loss": 5.083,
|
|
"mean_token_accuracy": 0.1862384185194969,
|
|
"num_tokens": 6546938.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.345768594741822,
|
|
"epoch": 2.9665271966527196,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044268189253825397,
|
|
"loss": 5.0953,
|
|
"mean_token_accuracy": 0.19227417707443237,
|
|
"num_tokens": 6555750.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.304913187026978,
|
|
"epoch": 2.9707112970711296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004424665052893334,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.2031712591648102,
|
|
"num_tokens": 6565680.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.225789165496826,
|
|
"epoch": 2.97489539748954,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044225077342126935,
|
|
"loss": 4.9977,
|
|
"mean_token_accuracy": 0.20195454061031343,
|
|
"num_tokens": 6574595.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.409015083312989,
|
|
"epoch": 2.97907949790795,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004420346973780033,
|
|
"loss": 5.1238,
|
|
"mean_token_accuracy": 0.1943576768040657,
|
|
"num_tokens": 6583333.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.445632266998291,
|
|
"epoch": 2.98326359832636,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000441818277604185,
|
|
"loss": 5.1427,
|
|
"mean_token_accuracy": 0.2017063707113266,
|
|
"num_tokens": 6592252.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.295730066299439,
|
|
"epoch": 2.98744769874477,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004416015145451715,
|
|
"loss": 4.9823,
|
|
"mean_token_accuracy": 0.2015913248062134,
|
|
"num_tokens": 6600675.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.308580780029297,
|
|
"epoch": 2.99163179916318,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004413844086470264,
|
|
"loss": 5.0693,
|
|
"mean_token_accuracy": 0.19256574362516404,
|
|
"num_tokens": 6609722.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.392408800125122,
|
|
"epoch": 2.99581589958159,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004411669603565187,
|
|
"loss": 5.0731,
|
|
"mean_token_accuracy": 0.203379026055336,
|
|
"num_tokens": 6618415.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.3663756370544435,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000440949170121122,
|
|
"loss": 5.0834,
|
|
"mean_token_accuracy": 0.19409166723489762,
|
|
"num_tokens": 6627966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.370896720886231,
|
|
"epoch": 3.00418410041841,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044073103838901373,
|
|
"loss": 4.7587,
|
|
"mean_token_accuracy": 0.22468715012073517,
|
|
"num_tokens": 6636820.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.254427099227906,
|
|
"epoch": 3.00836820083682,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044051256560907374,
|
|
"loss": 4.784,
|
|
"mean_token_accuracy": 0.20854490101337433,
|
|
"num_tokens": 6645792.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.432565021514892,
|
|
"epoch": 3.01255230125523,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00044029375223088394,
|
|
"loss": 4.8881,
|
|
"mean_token_accuracy": 0.20862485319375992,
|
|
"num_tokens": 6654850.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.307573938369751,
|
|
"epoch": 3.01673640167364,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044007459870472706,
|
|
"loss": 4.8123,
|
|
"mean_token_accuracy": 0.21202803701162337,
|
|
"num_tokens": 6664524.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.235706901550293,
|
|
"epoch": 3.02092050209205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004398551054815859,
|
|
"loss": 4.7463,
|
|
"mean_token_accuracy": 0.21180724650621413,
|
|
"num_tokens": 6674330.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.335667514801026,
|
|
"epoch": 3.0251046025104604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004396352730131422,
|
|
"loss": 4.8498,
|
|
"mean_token_accuracy": 0.2125999927520752,
|
|
"num_tokens": 6683121.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.331827545166016,
|
|
"epoch": 3.0292887029288704,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004394151017517758,
|
|
"loss": 4.7938,
|
|
"mean_token_accuracy": 0.21797282248735428,
|
|
"num_tokens": 6692566.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.347726488113404,
|
|
"epoch": 3.0334728033472804,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004391945921505638,
|
|
"loss": 4.8774,
|
|
"mean_token_accuracy": 0.21223192065954208,
|
|
"num_tokens": 6701577.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.228334426879883,
|
|
"epoch": 3.0376569037656904,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043897374466327944,
|
|
"loss": 4.7415,
|
|
"mean_token_accuracy": 0.2243683472275734,
|
|
"num_tokens": 6710672.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.251773548126221,
|
|
"epoch": 3.0418410041841004,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043875255974439145,
|
|
"loss": 4.7548,
|
|
"mean_token_accuracy": 0.21973217874765397,
|
|
"num_tokens": 6719412.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.2361396789550785,
|
|
"epoch": 3.0460251046025104,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004385310378490627,
|
|
"loss": 4.7453,
|
|
"mean_token_accuracy": 0.2163161501288414,
|
|
"num_tokens": 6729169.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.27430362701416,
|
|
"epoch": 3.0502092050209204,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043830917943314977,
|
|
"loss": 4.7056,
|
|
"mean_token_accuracy": 0.22477542906999587,
|
|
"num_tokens": 6738549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.226499223709107,
|
|
"epoch": 3.0543933054393304,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043808698495320154,
|
|
"loss": 4.7545,
|
|
"mean_token_accuracy": 0.21589447259902955,
|
|
"num_tokens": 6747909.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.217711305618286,
|
|
"epoch": 3.0585774058577404,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043786445486645846,
|
|
"loss": 4.8229,
|
|
"mean_token_accuracy": 0.21544221937656402,
|
|
"num_tokens": 6757125.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.30985426902771,
|
|
"epoch": 3.062761506276151,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004376415896308518,
|
|
"loss": 4.7936,
|
|
"mean_token_accuracy": 0.2216247320175171,
|
|
"num_tokens": 6765774.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.31474347114563,
|
|
"epoch": 3.066945606694561,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043741838970500234,
|
|
"loss": 4.8441,
|
|
"mean_token_accuracy": 0.21821288168430328,
|
|
"num_tokens": 6774980.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.308967399597168,
|
|
"epoch": 3.071129707112971,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004371948555482196,
|
|
"loss": 4.8463,
|
|
"mean_token_accuracy": 0.20979633927345276,
|
|
"num_tokens": 6784187.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.26036639213562,
|
|
"epoch": 3.075313807531381,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000436970987620501,
|
|
"loss": 4.8534,
|
|
"mean_token_accuracy": 0.21231846660375595,
|
|
"num_tokens": 6792953.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.27310733795166,
|
|
"epoch": 3.079497907949791,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043674678638253066,
|
|
"loss": 4.728,
|
|
"mean_token_accuracy": 0.22622442245483398,
|
|
"num_tokens": 6803004.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.257822132110595,
|
|
"epoch": 3.083682008368201,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004365222522956788,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.2126663088798523,
|
|
"num_tokens": 6811883.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.2302734375,
|
|
"epoch": 3.087866108786611,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004362973858220004,
|
|
"loss": 4.8007,
|
|
"mean_token_accuracy": 0.2243728145956993,
|
|
"num_tokens": 6820327.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.244401359558106,
|
|
"epoch": 3.092050209205021,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043607218742423447,
|
|
"loss": 4.8155,
|
|
"mean_token_accuracy": 0.21840747594833373,
|
|
"num_tokens": 6829412.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.2269069194793705,
|
|
"epoch": 3.096234309623431,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043584665756580323,
|
|
"loss": 4.7863,
|
|
"mean_token_accuracy": 0.2162407010793686,
|
|
"num_tokens": 6838997.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.316408300399781,
|
|
"epoch": 3.100418410041841,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004356207967108106,
|
|
"loss": 4.846,
|
|
"mean_token_accuracy": 0.2081720322370529,
|
|
"num_tokens": 6848957.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.294563817977905,
|
|
"epoch": 3.104602510460251,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004353946053240422,
|
|
"loss": 4.8863,
|
|
"mean_token_accuracy": 0.20891690105199814,
|
|
"num_tokens": 6859688.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.328656625747681,
|
|
"epoch": 3.1087866108786613,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004351680838709635,
|
|
"loss": 4.9089,
|
|
"mean_token_accuracy": 0.20370992869138718,
|
|
"num_tokens": 6869149.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.365725517272949,
|
|
"epoch": 3.1129707112970713,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043494123281771915,
|
|
"loss": 4.8916,
|
|
"mean_token_accuracy": 0.2070358246564865,
|
|
"num_tokens": 6879432.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.222113180160522,
|
|
"epoch": 3.1171548117154813,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043471405263113214,
|
|
"loss": 4.8473,
|
|
"mean_token_accuracy": 0.21447400897741317,
|
|
"num_tokens": 6888398.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.291485834121704,
|
|
"epoch": 3.1213389121338913,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004344865437787027,
|
|
"loss": 4.7762,
|
|
"mean_token_accuracy": 0.2152193993330002,
|
|
"num_tokens": 6898082.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.209522104263305,
|
|
"epoch": 3.1255230125523012,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043425870672860774,
|
|
"loss": 4.7545,
|
|
"mean_token_accuracy": 0.21814098060131074,
|
|
"num_tokens": 6907693.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.319671583175659,
|
|
"epoch": 3.1297071129707112,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004340305419496989,
|
|
"loss": 4.9537,
|
|
"mean_token_accuracy": 0.20594608932733535,
|
|
"num_tokens": 6917095.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.3316257953643795,
|
|
"epoch": 3.1338912133891212,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004338020499115029,
|
|
"loss": 4.8867,
|
|
"mean_token_accuracy": 0.21243216544389726,
|
|
"num_tokens": 6926952.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.276588678359985,
|
|
"epoch": 3.1380753138075312,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004335732310842196,
|
|
"loss": 4.8222,
|
|
"mean_token_accuracy": 0.20978376418352127,
|
|
"num_tokens": 6936704.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.222697925567627,
|
|
"epoch": 3.1422594142259412,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004333440859387212,
|
|
"loss": 4.839,
|
|
"mean_token_accuracy": 0.2161761537194252,
|
|
"num_tokens": 6946614.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.37057638168335,
|
|
"epoch": 3.1464435146443517,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043311461494655165,
|
|
"loss": 4.8582,
|
|
"mean_token_accuracy": 0.2197693705558777,
|
|
"num_tokens": 6955980.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.325770902633667,
|
|
"epoch": 3.1506276150627617,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004328848185799254,
|
|
"loss": 4.8273,
|
|
"mean_token_accuracy": 0.213053897023201,
|
|
"num_tokens": 6964108.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.260280275344849,
|
|
"epoch": 3.1548117154811717,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004326546973117264,
|
|
"loss": 4.8239,
|
|
"mean_token_accuracy": 0.22190679758787155,
|
|
"num_tokens": 6972895.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.254884576797485,
|
|
"epoch": 3.1589958158995817,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043242425161550723,
|
|
"loss": 4.8468,
|
|
"mean_token_accuracy": 0.2187548264861107,
|
|
"num_tokens": 6982222.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.332245588302612,
|
|
"epoch": 3.1631799163179917,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043219348196548807,
|
|
"loss": 4.8362,
|
|
"mean_token_accuracy": 0.2134234443306923,
|
|
"num_tokens": 6990820.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.275543308258056,
|
|
"epoch": 3.1673640167364017,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004319623888365557,
|
|
"loss": 4.8721,
|
|
"mean_token_accuracy": 0.21123863160610198,
|
|
"num_tokens": 7000846.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.202007675170899,
|
|
"epoch": 3.1715481171548117,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043173097270426275,
|
|
"loss": 4.788,
|
|
"mean_token_accuracy": 0.21145834624767304,
|
|
"num_tokens": 7009086.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.419174480438232,
|
|
"epoch": 3.1757322175732217,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043149923404482634,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.2108924940228462,
|
|
"num_tokens": 7017752.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.2178784847259525,
|
|
"epoch": 3.1799163179916317,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043126717333512733,
|
|
"loss": 4.8523,
|
|
"mean_token_accuracy": 0.2184145227074623,
|
|
"num_tokens": 7026535.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.2732611179351805,
|
|
"epoch": 3.1841004184100417,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004310347910527094,
|
|
"loss": 4.8931,
|
|
"mean_token_accuracy": 0.2162533685564995,
|
|
"num_tokens": 7035462.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.297032070159912,
|
|
"epoch": 3.1882845188284517,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004308020876757778,
|
|
"loss": 4.8061,
|
|
"mean_token_accuracy": 0.21692827492952346,
|
|
"num_tokens": 7043978.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.247585582733154,
|
|
"epoch": 3.192468619246862,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000430569063683199,
|
|
"loss": 4.8158,
|
|
"mean_token_accuracy": 0.21437426954507827,
|
|
"num_tokens": 7053757.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.179677152633667,
|
|
"epoch": 3.196652719665272,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004303357195544986,
|
|
"loss": 4.7438,
|
|
"mean_token_accuracy": 0.21467897593975066,
|
|
"num_tokens": 7062491.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.2135216236114506,
|
|
"epoch": 3.200836820083682,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004301020557698615,
|
|
"loss": 4.7805,
|
|
"mean_token_accuracy": 0.2226917862892151,
|
|
"num_tokens": 7072157.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.251100969314575,
|
|
"epoch": 3.205020920502092,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042986807281013013,
|
|
"loss": 4.7528,
|
|
"mean_token_accuracy": 0.2188943773508072,
|
|
"num_tokens": 7080631.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.299303436279297,
|
|
"epoch": 3.209205020920502,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004296337711568037,
|
|
"loss": 4.8286,
|
|
"mean_token_accuracy": 0.22220218032598496,
|
|
"num_tokens": 7089944.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.2891419410705565,
|
|
"epoch": 3.213389121338912,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042939915129203745,
|
|
"loss": 4.8803,
|
|
"mean_token_accuracy": 0.21736244559288026,
|
|
"num_tokens": 7099190.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.124616384506226,
|
|
"epoch": 3.217573221757322,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004291642136986413,
|
|
"loss": 4.8362,
|
|
"mean_token_accuracy": 0.20817089080810547,
|
|
"num_tokens": 7107944.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.24336609840393,
|
|
"epoch": 3.221757322175732,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000428928958860079,
|
|
"loss": 4.679,
|
|
"mean_token_accuracy": 0.22218458652496337,
|
|
"num_tokens": 7116953.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.174569749832154,
|
|
"epoch": 3.225941422594142,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004286933872604673,
|
|
"loss": 4.743,
|
|
"mean_token_accuracy": 0.22391242235898973,
|
|
"num_tokens": 7126236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.257703495025635,
|
|
"epoch": 3.2301255230125525,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004284574993845746,
|
|
"loss": 4.8164,
|
|
"mean_token_accuracy": 0.217802557349205,
|
|
"num_tokens": 7136286.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.2871177196502686,
|
|
"epoch": 3.2343096234309625,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042822129571782025,
|
|
"loss": 4.8679,
|
|
"mean_token_accuracy": 0.21115810722112655,
|
|
"num_tokens": 7145551.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.275178241729736,
|
|
"epoch": 3.2384937238493725,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042798477674627343,
|
|
"loss": 4.8351,
|
|
"mean_token_accuracy": 0.21977209597826003,
|
|
"num_tokens": 7155486.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.251350736618042,
|
|
"epoch": 3.2426778242677825,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042774794295665213,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.2152801752090454,
|
|
"num_tokens": 7163984.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.232707452774048,
|
|
"epoch": 3.2468619246861925,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004275107948363223,
|
|
"loss": 4.7732,
|
|
"mean_token_accuracy": 0.21727198660373687,
|
|
"num_tokens": 7172854.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.171909189224243,
|
|
"epoch": 3.2510460251046025,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004272733328732965,
|
|
"loss": 4.7955,
|
|
"mean_token_accuracy": 0.21165214329957963,
|
|
"num_tokens": 7182202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.286756610870361,
|
|
"epoch": 3.2552301255230125,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042703555755623337,
|
|
"loss": 4.9027,
|
|
"mean_token_accuracy": 0.20935914516448975,
|
|
"num_tokens": 7192062.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.383759784698486,
|
|
"epoch": 3.2594142259414225,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004267974693744362,
|
|
"loss": 4.8957,
|
|
"mean_token_accuracy": 0.21058278232812883,
|
|
"num_tokens": 7201766.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.243042659759522,
|
|
"epoch": 3.2635983263598325,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004265590688178523,
|
|
"loss": 4.8277,
|
|
"mean_token_accuracy": 0.20863327085971833,
|
|
"num_tokens": 7211361.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.2302161693573,
|
|
"epoch": 3.2677824267782425,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004263203563770716,
|
|
"loss": 4.7923,
|
|
"mean_token_accuracy": 0.2196310967206955,
|
|
"num_tokens": 7220221.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.285264873504639,
|
|
"epoch": 3.2719665271966525,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000426081332543326,
|
|
"loss": 4.9064,
|
|
"mean_token_accuracy": 0.20687966793775558,
|
|
"num_tokens": 7230134.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.148596382141113,
|
|
"epoch": 3.276150627615063,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004258419978084881,
|
|
"loss": 4.7173,
|
|
"mean_token_accuracy": 0.22078851312398912,
|
|
"num_tokens": 7238713.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.304853248596191,
|
|
"epoch": 3.280334728033473,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004256023526650702,
|
|
"loss": 4.8434,
|
|
"mean_token_accuracy": 0.21353326439857484,
|
|
"num_tokens": 7247603.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.2230082035064695,
|
|
"epoch": 3.284518828451883,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004253623976062236,
|
|
"loss": 4.8746,
|
|
"mean_token_accuracy": 0.2149614214897156,
|
|
"num_tokens": 7257452.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.408122587203979,
|
|
"epoch": 3.288702928870293,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004251221331257372,
|
|
"loss": 4.8759,
|
|
"mean_token_accuracy": 0.21849010437726973,
|
|
"num_tokens": 7266401.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.225405597686768,
|
|
"epoch": 3.292887029288703,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004248815597180366,
|
|
"loss": 4.8294,
|
|
"mean_token_accuracy": 0.21766505837440492,
|
|
"num_tokens": 7275836.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.290088272094726,
|
|
"epoch": 3.297071129707113,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042464067787818333,
|
|
"loss": 4.8757,
|
|
"mean_token_accuracy": 0.2104683741927147,
|
|
"num_tokens": 7284854.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.264577007293701,
|
|
"epoch": 3.301255230125523,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004243994881018735,
|
|
"loss": 4.8291,
|
|
"mean_token_accuracy": 0.21094449758529663,
|
|
"num_tokens": 7293421.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.222991561889648,
|
|
"epoch": 3.305439330543933,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004241579908854367,
|
|
"loss": 4.7792,
|
|
"mean_token_accuracy": 0.2157828614115715,
|
|
"num_tokens": 7302184.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.304621648788452,
|
|
"epoch": 3.309623430962343,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042391618672583546,
|
|
"loss": 4.9391,
|
|
"mean_token_accuracy": 0.21059073358774186,
|
|
"num_tokens": 7311596.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.285192489624023,
|
|
"epoch": 3.3138075313807533,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042367407612066383,
|
|
"loss": 4.7555,
|
|
"mean_token_accuracy": 0.2206650584936142,
|
|
"num_tokens": 7319866.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.2536461353302,
|
|
"epoch": 3.3179916317991633,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004234316595681468,
|
|
"loss": 4.8059,
|
|
"mean_token_accuracy": 0.2173109918832779,
|
|
"num_tokens": 7329327.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.340837383270264,
|
|
"epoch": 3.3221757322175733,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004231889375671382,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.20075691789388656,
|
|
"num_tokens": 7339717.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.2716363906860355,
|
|
"epoch": 3.3263598326359833,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042294591061712124,
|
|
"loss": 4.8156,
|
|
"mean_token_accuracy": 0.2170746624469757,
|
|
"num_tokens": 7348850.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.2059375762939455,
|
|
"epoch": 3.3305439330543933,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042270257921820623,
|
|
"loss": 4.773,
|
|
"mean_token_accuracy": 0.22047581821680068,
|
|
"num_tokens": 7357081.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.1577067375183105,
|
|
"epoch": 3.3347280334728033,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00042245894387112987,
|
|
"loss": 4.7889,
|
|
"mean_token_accuracy": 0.22042840868234634,
|
|
"num_tokens": 7365762.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.207553815841675,
|
|
"epoch": 3.3389121338912133,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004222150050772548,
|
|
"loss": 4.8533,
|
|
"mean_token_accuracy": 0.21571688801050187,
|
|
"num_tokens": 7374870.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.2268860816955565,
|
|
"epoch": 3.3430962343096233,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004219707633385677,
|
|
"loss": 4.7227,
|
|
"mean_token_accuracy": 0.22404564470052718,
|
|
"num_tokens": 7383836.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.230204057693482,
|
|
"epoch": 3.3472803347280333,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042172621915767886,
|
|
"loss": 4.8299,
|
|
"mean_token_accuracy": 0.21861498802900314,
|
|
"num_tokens": 7394059.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3472803347280333,
|
|
"eval_entropy": 4.971586591139405,
|
|
"eval_loss": 5.366418838500977,
|
|
"eval_mean_token_accuracy": 0.19403364545911367,
|
|
"eval_num_tokens": 7394059.0,
|
|
"eval_runtime": 2.1052,
|
|
"eval_samples_per_second": 1855.881,
|
|
"eval_steps_per_second": 232.282,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.152875566482544,
|
|
"epoch": 3.3514644351464433,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00042148137303782077,
|
|
"loss": 4.7726,
|
|
"mean_token_accuracy": 0.22096417248249053,
|
|
"num_tokens": 7402947.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.254004955291748,
|
|
"epoch": 3.3556485355648533,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004212362254828475,
|
|
"loss": 4.8436,
|
|
"mean_token_accuracy": 0.21177827417850495,
|
|
"num_tokens": 7413430.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.275895261764527,
|
|
"epoch": 3.3598326359832638,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042099077699723326,
|
|
"loss": 4.8485,
|
|
"mean_token_accuracy": 0.21575509160757064,
|
|
"num_tokens": 7423239.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.270434761047364,
|
|
"epoch": 3.3640167364016738,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004207450280860716,
|
|
"loss": 4.8449,
|
|
"mean_token_accuracy": 0.21819503903388976,
|
|
"num_tokens": 7432519.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.194208097457886,
|
|
"epoch": 3.3682008368200838,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004204989792550742,
|
|
"loss": 4.8136,
|
|
"mean_token_accuracy": 0.21632040292024612,
|
|
"num_tokens": 7442089.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.255561256408692,
|
|
"epoch": 3.3723849372384938,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00042025263101057016,
|
|
"loss": 4.8365,
|
|
"mean_token_accuracy": 0.21220530569553375,
|
|
"num_tokens": 7451059.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.228648376464844,
|
|
"epoch": 3.3765690376569037,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042000598385950427,
|
|
"loss": 4.7785,
|
|
"mean_token_accuracy": 0.21426960527896882,
|
|
"num_tokens": 7458960.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.238389015197754,
|
|
"epoch": 3.3807531380753137,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000419759038309437,
|
|
"loss": 4.9554,
|
|
"mean_token_accuracy": 0.20450644344091415,
|
|
"num_tokens": 7468863.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.297040557861328,
|
|
"epoch": 3.3849372384937237,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00041951179486854253,
|
|
"loss": 4.9125,
|
|
"mean_token_accuracy": 0.20795511603355407,
|
|
"num_tokens": 7477204.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.23099045753479,
|
|
"epoch": 3.3891213389121337,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000419264254045608,
|
|
"loss": 4.7543,
|
|
"mean_token_accuracy": 0.21470353305339812,
|
|
"num_tokens": 7487367.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.1249110221862795,
|
|
"epoch": 3.393305439330544,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004190164163500326,
|
|
"loss": 4.7321,
|
|
"mean_token_accuracy": 0.22717221677303315,
|
|
"num_tokens": 7495808.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.209413242340088,
|
|
"epoch": 3.397489539748954,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004187682822918266,
|
|
"loss": 4.8857,
|
|
"mean_token_accuracy": 0.19950032383203506,
|
|
"num_tokens": 7506119.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.236322975158691,
|
|
"epoch": 3.401673640167364,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041851985238161,
|
|
"loss": 4.8465,
|
|
"mean_token_accuracy": 0.21434216797351838,
|
|
"num_tokens": 7514931.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.131845760345459,
|
|
"epoch": 3.405857740585774,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041827112713061155,
|
|
"loss": 4.6607,
|
|
"mean_token_accuracy": 0.2306380018591881,
|
|
"num_tokens": 7523753.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.22949481010437,
|
|
"epoch": 3.410041841004184,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00041802210705066775,
|
|
"loss": 4.7402,
|
|
"mean_token_accuracy": 0.22057479172945021,
|
|
"num_tokens": 7532447.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.168293619155884,
|
|
"epoch": 3.414225941422594,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000417772792654222,
|
|
"loss": 4.8957,
|
|
"mean_token_accuracy": 0.20950355529785156,
|
|
"num_tokens": 7541388.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.260436153411865,
|
|
"epoch": 3.418410041841004,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041752318445432323,
|
|
"loss": 4.8958,
|
|
"mean_token_accuracy": 0.20558364391326905,
|
|
"num_tokens": 7550422.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.244349431991577,
|
|
"epoch": 3.422594142259414,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000417273282964625,
|
|
"loss": 4.7569,
|
|
"mean_token_accuracy": 0.2244752451777458,
|
|
"num_tokens": 7559522.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.245428943634034,
|
|
"epoch": 3.426778242677824,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004170230886993844,
|
|
"loss": 4.8686,
|
|
"mean_token_accuracy": 0.2094870388507843,
|
|
"num_tokens": 7568785.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.211488533020019,
|
|
"epoch": 3.430962343096234,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041677260217346093,
|
|
"loss": 4.7901,
|
|
"mean_token_accuracy": 0.21364631354808808,
|
|
"num_tokens": 7579013.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.203430271148681,
|
|
"epoch": 3.435146443514644,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041652182390231556,
|
|
"loss": 4.8118,
|
|
"mean_token_accuracy": 0.21443704813718795,
|
|
"num_tokens": 7588577.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.156320428848266,
|
|
"epoch": 3.439330543933054,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041627075440200977,
|
|
"loss": 4.7486,
|
|
"mean_token_accuracy": 0.22256388515233994,
|
|
"num_tokens": 7597702.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.1995807647705075,
|
|
"epoch": 3.4435146443514646,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004160193941892043,
|
|
"loss": 4.8124,
|
|
"mean_token_accuracy": 0.21778603196144103,
|
|
"num_tokens": 7607156.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.257018232345581,
|
|
"epoch": 3.4476987447698746,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004157677437811578,
|
|
"loss": 4.8902,
|
|
"mean_token_accuracy": 0.22197027802467345,
|
|
"num_tokens": 7615936.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.288289403915405,
|
|
"epoch": 3.4518828451882846,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004155158036957265,
|
|
"loss": 4.8654,
|
|
"mean_token_accuracy": 0.2177944153547287,
|
|
"num_tokens": 7626568.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.2945146560668945,
|
|
"epoch": 3.4560669456066946,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004152635744513626,
|
|
"loss": 4.8114,
|
|
"mean_token_accuracy": 0.21972503811120986,
|
|
"num_tokens": 7635102.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.17290244102478,
|
|
"epoch": 3.4602510460251046,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004150110565671133,
|
|
"loss": 4.7353,
|
|
"mean_token_accuracy": 0.22908539175987244,
|
|
"num_tokens": 7644204.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.229699277877808,
|
|
"epoch": 3.4644351464435146,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00041475825056261985,
|
|
"loss": 4.8564,
|
|
"mean_token_accuracy": 0.20910507291555405,
|
|
"num_tokens": 7653585.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.211961841583252,
|
|
"epoch": 3.4686192468619246,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004145051569581162,
|
|
"loss": 4.8442,
|
|
"mean_token_accuracy": 0.20911729335784912,
|
|
"num_tokens": 7662734.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.261643981933593,
|
|
"epoch": 3.4728033472803346,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004142517762744286,
|
|
"loss": 4.8091,
|
|
"mean_token_accuracy": 0.21733799427747727,
|
|
"num_tokens": 7671682.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.304840517044068,
|
|
"epoch": 3.476987447698745,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004139981090329735,
|
|
"loss": 4.9181,
|
|
"mean_token_accuracy": 0.212834532558918,
|
|
"num_tokens": 7681551.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.275520896911621,
|
|
"epoch": 3.481171548117155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004137441557557575,
|
|
"loss": 4.856,
|
|
"mean_token_accuracy": 0.21502091884613037,
|
|
"num_tokens": 7690943.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.196260213851929,
|
|
"epoch": 3.485355648535565,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041348991696537533,
|
|
"loss": 4.7653,
|
|
"mean_token_accuracy": 0.21998494118452072,
|
|
"num_tokens": 7700376.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.279877614974976,
|
|
"epoch": 3.489539748953975,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004132353931850098,
|
|
"loss": 4.8555,
|
|
"mean_token_accuracy": 0.20926144272089003,
|
|
"num_tokens": 7709845.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.302618789672851,
|
|
"epoch": 3.493723849372385,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004129805849384299,
|
|
"loss": 4.9579,
|
|
"mean_token_accuracy": 0.2061849594116211,
|
|
"num_tokens": 7719908.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.235215520858764,
|
|
"epoch": 3.497907949790795,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004127254927499898,
|
|
"loss": 4.8711,
|
|
"mean_token_accuracy": 0.21028606593608856,
|
|
"num_tokens": 7729543.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.285682678222656,
|
|
"epoch": 3.502092050209205,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004124701171446286,
|
|
"loss": 4.8933,
|
|
"mean_token_accuracy": 0.21481493413448333,
|
|
"num_tokens": 7739371.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.243874883651733,
|
|
"epoch": 3.506276150627615,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00041221445864786794,
|
|
"loss": 4.8069,
|
|
"mean_token_accuracy": 0.21638569682836534,
|
|
"num_tokens": 7748027.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.147857761383056,
|
|
"epoch": 3.510460251046025,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000411958517785812,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.22171512246131897,
|
|
"num_tokens": 7756588.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.18248405456543,
|
|
"epoch": 3.514644351464435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041170229508514584,
|
|
"loss": 4.7233,
|
|
"mean_token_accuracy": 0.21787351071834565,
|
|
"num_tokens": 7765222.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.226833248138428,
|
|
"epoch": 3.518828451882845,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004114457910731347,
|
|
"loss": 4.8324,
|
|
"mean_token_accuracy": 0.21579353958368303,
|
|
"num_tokens": 7774139.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.179098606109619,
|
|
"epoch": 3.523012552301255,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004111890062776226,
|
|
"loss": 4.7128,
|
|
"mean_token_accuracy": 0.21928198188543319,
|
|
"num_tokens": 7783597.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.095175504684448,
|
|
"epoch": 3.5271966527196654,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004109319412270314,
|
|
"loss": 4.8108,
|
|
"mean_token_accuracy": 0.2205233097076416,
|
|
"num_tokens": 7792711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.322521829605103,
|
|
"epoch": 3.5313807531380754,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004106745964503594,
|
|
"loss": 4.8941,
|
|
"mean_token_accuracy": 0.21111726611852646,
|
|
"num_tokens": 7801878.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.238970470428467,
|
|
"epoch": 3.5355648535564854,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000410416972477181,
|
|
"loss": 4.7823,
|
|
"mean_token_accuracy": 0.22111913859844207,
|
|
"num_tokens": 7810484.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.252415418624878,
|
|
"epoch": 3.5397489539748954,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00041015906983764495,
|
|
"loss": 4.8827,
|
|
"mean_token_accuracy": 0.21148828715085982,
|
|
"num_tokens": 7820521.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.229612255096436,
|
|
"epoch": 3.5439330543933054,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00040990088906247326,
|
|
"loss": 4.8019,
|
|
"mean_token_accuracy": 0.21767462491989137,
|
|
"num_tokens": 7829218.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.220323085784912,
|
|
"epoch": 3.5481171548117154,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004096424306829605,
|
|
"loss": 4.818,
|
|
"mean_token_accuracy": 0.21166364401578902,
|
|
"num_tokens": 7837795.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.240948295593261,
|
|
"epoch": 3.5523012552301254,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004093836952309726,
|
|
"loss": 4.83,
|
|
"mean_token_accuracy": 0.21202095597982407,
|
|
"num_tokens": 7846973.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.29024305343628,
|
|
"epoch": 3.5564853556485354,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004091246832389453,
|
|
"loss": 4.8777,
|
|
"mean_token_accuracy": 0.2121277019381523,
|
|
"num_tokens": 7856627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.1871363639831545,
|
|
"epoch": 3.560669456066946,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004088653952398837,
|
|
"loss": 4.7838,
|
|
"mean_token_accuracy": 0.22284585982561111,
|
|
"num_tokens": 7866308.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.192884349822998,
|
|
"epoch": 3.564853556485356,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00040860583176736097,
|
|
"loss": 4.784,
|
|
"mean_token_accuracy": 0.2220319151878357,
|
|
"num_tokens": 7875350.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.172300100326538,
|
|
"epoch": 3.569037656903766,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004083459933555169,
|
|
"loss": 4.7982,
|
|
"mean_token_accuracy": 0.22210454791784287,
|
|
"num_tokens": 7884276.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.242338466644287,
|
|
"epoch": 3.573221757322176,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004080858805390571,
|
|
"loss": 4.8572,
|
|
"mean_token_accuracy": 0.21125928610563277,
|
|
"num_tokens": 7893836.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.257506799697876,
|
|
"epoch": 3.577405857740586,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000407825493853252,
|
|
"loss": 4.8449,
|
|
"mean_token_accuracy": 0.22228922545909882,
|
|
"num_tokens": 7903249.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.221683025360107,
|
|
"epoch": 3.581589958158996,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040756483383393556,
|
|
"loss": 4.8885,
|
|
"mean_token_accuracy": 0.20869875848293304,
|
|
"num_tokens": 7912217.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.2422223567962645,
|
|
"epoch": 3.585774058577406,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004073039010175041,
|
|
"loss": 4.927,
|
|
"mean_token_accuracy": 0.20969937443733216,
|
|
"num_tokens": 7922241.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.264800262451172,
|
|
"epoch": 3.589958158995816,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040704269594091556,
|
|
"loss": 4.8312,
|
|
"mean_token_accuracy": 0.2115737795829773,
|
|
"num_tokens": 7931866.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.273177003860473,
|
|
"epoch": 3.594142259414226,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004067812191416878,
|
|
"loss": 4.8683,
|
|
"mean_token_accuracy": 0.21232162564992904,
|
|
"num_tokens": 7941787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.2348886013031,
|
|
"epoch": 3.598326359832636,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004065194711578982,
|
|
"loss": 4.8728,
|
|
"mean_token_accuracy": 0.20931871831417084,
|
|
"num_tokens": 7950832.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.261820459365845,
|
|
"epoch": 3.602510460251046,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004062574525281819,
|
|
"loss": 4.8811,
|
|
"mean_token_accuracy": 0.21396803557872773,
|
|
"num_tokens": 7960059.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.26634030342102,
|
|
"epoch": 3.606694560669456,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00040599516379173113,
|
|
"loss": 4.8443,
|
|
"mean_token_accuracy": 0.20981826037168502,
|
|
"num_tokens": 7969411.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.215761470794678,
|
|
"epoch": 3.6108786610878663,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00040573260548829417,
|
|
"loss": 4.8602,
|
|
"mean_token_accuracy": 0.21079269796609879,
|
|
"num_tokens": 7978446.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.128013277053833,
|
|
"epoch": 3.6150627615062763,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004054697781581734,
|
|
"loss": 4.749,
|
|
"mean_token_accuracy": 0.21835661083459854,
|
|
"num_tokens": 7987117.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.244450426101684,
|
|
"epoch": 3.6192468619246863,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004052066823422255,
|
|
"loss": 4.8252,
|
|
"mean_token_accuracy": 0.21355912536382676,
|
|
"num_tokens": 7995989.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.18575873374939,
|
|
"epoch": 3.6234309623430963,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004049433185818593,
|
|
"loss": 4.8783,
|
|
"mean_token_accuracy": 0.20862276703119279,
|
|
"num_tokens": 8005486.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.179786014556885,
|
|
"epoch": 3.6276150627615062,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040467968741903507,
|
|
"loss": 4.7813,
|
|
"mean_token_accuracy": 0.21884793639183045,
|
|
"num_tokens": 8014853.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.278134679794311,
|
|
"epoch": 3.6317991631799162,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004044157893962633,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.21482334583997725,
|
|
"num_tokens": 8023931.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.244082689285278,
|
|
"epoch": 3.6359832635983262,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004041516250566037,
|
|
"loss": 4.8339,
|
|
"mean_token_accuracy": 0.21539998501539231,
|
|
"num_tokens": 8032659.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.206443500518799,
|
|
"epoch": 3.6401673640167362,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040388719494366404,
|
|
"loss": 4.9199,
|
|
"mean_token_accuracy": 0.21264792531728743,
|
|
"num_tokens": 8041857.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.316039562225342,
|
|
"epoch": 3.6443514644351467,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040362249960159904,
|
|
"loss": 4.9266,
|
|
"mean_token_accuracy": 0.20998856872320176,
|
|
"num_tokens": 8051186.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.23778281211853,
|
|
"epoch": 3.6485355648535567,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040335753957510906,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.20667103230953215,
|
|
"num_tokens": 8060396.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.182791328430175,
|
|
"epoch": 3.6527196652719667,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040309231540943937,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.21902086585760117,
|
|
"num_tokens": 8068106.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.245421934127807,
|
|
"epoch": 3.6569037656903767,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004028268276503784,
|
|
"loss": 4.8109,
|
|
"mean_token_accuracy": 0.2120743915438652,
|
|
"num_tokens": 8077878.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.303288507461548,
|
|
"epoch": 3.6610878661087867,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004025610768442576,
|
|
"loss": 4.9286,
|
|
"mean_token_accuracy": 0.20912135243415833,
|
|
"num_tokens": 8086968.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.195829439163208,
|
|
"epoch": 3.6652719665271967,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004022950635379492,
|
|
"loss": 4.7446,
|
|
"mean_token_accuracy": 0.22568707764148713,
|
|
"num_tokens": 8095999.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.133271932601929,
|
|
"epoch": 3.6694560669456067,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040202878827886587,
|
|
"loss": 4.7453,
|
|
"mean_token_accuracy": 0.2247105523943901,
|
|
"num_tokens": 8105376.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.246076059341431,
|
|
"epoch": 3.6736401673640167,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004017622516149592,
|
|
"loss": 4.8595,
|
|
"mean_token_accuracy": 0.21905912309885026,
|
|
"num_tokens": 8115024.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.237581729888916,
|
|
"epoch": 3.6778242677824267,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004014954540947189,
|
|
"loss": 4.864,
|
|
"mean_token_accuracy": 0.21400754451751708,
|
|
"num_tokens": 8124306.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.171728277206421,
|
|
"epoch": 3.6820083682008367,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00040122839626717137,
|
|
"loss": 4.8435,
|
|
"mean_token_accuracy": 0.21770024299621582,
|
|
"num_tokens": 8133151.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.230675077438354,
|
|
"epoch": 3.6861924686192467,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004009610786818786,
|
|
"loss": 4.8633,
|
|
"mean_token_accuracy": 0.21793696731328965,
|
|
"num_tokens": 8141745.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.143944263458252,
|
|
"epoch": 3.6903765690376567,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040069350188893735,
|
|
"loss": 4.7084,
|
|
"mean_token_accuracy": 0.22158368974924086,
|
|
"num_tokens": 8150884.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.2093270778656,
|
|
"epoch": 3.694560669456067,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040042566643897745,
|
|
"loss": 4.8179,
|
|
"mean_token_accuracy": 0.21036275178194047,
|
|
"num_tokens": 8160138.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.2616699695587155,
|
|
"epoch": 3.698744769874477,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004001575728831613,
|
|
"loss": 4.8835,
|
|
"mean_token_accuracy": 0.2155214309692383,
|
|
"num_tokens": 8169655.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.275929355621338,
|
|
"epoch": 3.702928870292887,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003998892217731824,
|
|
"loss": 4.8843,
|
|
"mean_token_accuracy": 0.2161941260099411,
|
|
"num_tokens": 8178846.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.193227338790893,
|
|
"epoch": 3.707112970711297,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000399620613661264,
|
|
"loss": 4.7918,
|
|
"mean_token_accuracy": 0.21546046286821366,
|
|
"num_tokens": 8187525.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.204625034332276,
|
|
"epoch": 3.711297071129707,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003993517491001586,
|
|
"loss": 4.8834,
|
|
"mean_token_accuracy": 0.20841171145439147,
|
|
"num_tokens": 8196855.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.088031864166259,
|
|
"epoch": 3.715481171548117,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003990826286431462,
|
|
"loss": 4.7522,
|
|
"mean_token_accuracy": 0.21392904669046403,
|
|
"num_tokens": 8205530.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.234610843658447,
|
|
"epoch": 3.719665271966527,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003988132528440333,
|
|
"loss": 4.8455,
|
|
"mean_token_accuracy": 0.21409202367067337,
|
|
"num_tokens": 8215066.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.275286769866943,
|
|
"epoch": 3.723849372384937,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003985436222571521,
|
|
"loss": 4.84,
|
|
"mean_token_accuracy": 0.21313633471727372,
|
|
"num_tokens": 8224585.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.260278415679932,
|
|
"epoch": 3.7280334728033475,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000398273737437359,
|
|
"loss": 4.9022,
|
|
"mean_token_accuracy": 0.2075318530201912,
|
|
"num_tokens": 8233334.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.255334043502808,
|
|
"epoch": 3.7322175732217575,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039800359894003355,
|
|
"loss": 4.8228,
|
|
"mean_token_accuracy": 0.22250812649726867,
|
|
"num_tokens": 8241636.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.244854402542114,
|
|
"epoch": 3.7364016736401675,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003977332073210774,
|
|
"loss": 4.8731,
|
|
"mean_token_accuracy": 0.21416753232479097,
|
|
"num_tokens": 8250273.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.236929941177368,
|
|
"epoch": 3.7405857740585775,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003974625631369129,
|
|
"loss": 4.9025,
|
|
"mean_token_accuracy": 0.20657675266265868,
|
|
"num_tokens": 8260349.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.33499665260315,
|
|
"epoch": 3.7447698744769875,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003971916669444824,
|
|
"loss": 4.9024,
|
|
"mean_token_accuracy": 0.21088040471076966,
|
|
"num_tokens": 8270505.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.285224437713623,
|
|
"epoch": 3.7489539748953975,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039692051930124663,
|
|
"loss": 4.8666,
|
|
"mean_token_accuracy": 0.21131369471549988,
|
|
"num_tokens": 8278651.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.196974754333496,
|
|
"epoch": 3.7531380753138075,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003966491207651838,
|
|
"loss": 4.814,
|
|
"mean_token_accuracy": 0.2202211230993271,
|
|
"num_tokens": 8287173.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.27840371131897,
|
|
"epoch": 3.7573221757322175,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039637747189478866,
|
|
"loss": 4.82,
|
|
"mean_token_accuracy": 0.21750265508890151,
|
|
"num_tokens": 8295785.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.239850568771362,
|
|
"epoch": 3.7615062761506275,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003961055732490708,
|
|
"loss": 4.9404,
|
|
"mean_token_accuracy": 0.2123316764831543,
|
|
"num_tokens": 8305534.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.2374858379364015,
|
|
"epoch": 3.7656903765690375,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00039583342538755374,
|
|
"loss": 4.8301,
|
|
"mean_token_accuracy": 0.21941172927618027,
|
|
"num_tokens": 8314232.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.7656903765690375,
|
|
"eval_entropy": 4.99278355132339,
|
|
"eval_loss": 5.324733734130859,
|
|
"eval_mean_token_accuracy": 0.19716157343672827,
|
|
"eval_num_tokens": 8314232.0,
|
|
"eval_runtime": 2.0874,
|
|
"eval_samples_per_second": 1871.693,
|
|
"eval_steps_per_second": 234.261,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.2630095958709715,
|
|
"epoch": 3.7698744769874475,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039556102887027423,
|
|
"loss": 4.8889,
|
|
"mean_token_accuracy": 0.20684399008750914,
|
|
"num_tokens": 8323230.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.281521368026733,
|
|
"epoch": 3.7740585774058575,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039528838425778053,
|
|
"loss": 4.9024,
|
|
"mean_token_accuracy": 0.21231130957603456,
|
|
"num_tokens": 8333072.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.2856419563293455,
|
|
"epoch": 3.778242677824268,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003950154921111313,
|
|
"loss": 4.8647,
|
|
"mean_token_accuracy": 0.20804665237665176,
|
|
"num_tokens": 8342953.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.206783485412598,
|
|
"epoch": 3.782426778242678,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039474235299189484,
|
|
"loss": 4.8657,
|
|
"mean_token_accuracy": 0.21000930219888686,
|
|
"num_tokens": 8351729.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.259337472915649,
|
|
"epoch": 3.786610878661088,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003944689674621474,
|
|
"loss": 4.9266,
|
|
"mean_token_accuracy": 0.21238088458776475,
|
|
"num_tokens": 8360929.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.347767543792725,
|
|
"epoch": 3.790794979079498,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039419533608447244,
|
|
"loss": 4.9032,
|
|
"mean_token_accuracy": 0.2159664049744606,
|
|
"num_tokens": 8369506.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.267509889602661,
|
|
"epoch": 3.794979079497908,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003939214594219595,
|
|
"loss": 4.8177,
|
|
"mean_token_accuracy": 0.21720889955759048,
|
|
"num_tokens": 8378583.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.256266355514526,
|
|
"epoch": 3.799163179916318,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039364733803820263,
|
|
"loss": 4.811,
|
|
"mean_token_accuracy": 0.21555519998073577,
|
|
"num_tokens": 8388078.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.128119087219238,
|
|
"epoch": 3.803347280334728,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003933729724972996,
|
|
"loss": 4.7989,
|
|
"mean_token_accuracy": 0.2176191434264183,
|
|
"num_tokens": 8397086.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.2004142761230465,
|
|
"epoch": 3.8075313807531384,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039309836336385044,
|
|
"loss": 4.7956,
|
|
"mean_token_accuracy": 0.23081605285406112,
|
|
"num_tokens": 8405621.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.292717409133911,
|
|
"epoch": 3.8117154811715483,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003928235112029568,
|
|
"loss": 4.8599,
|
|
"mean_token_accuracy": 0.21450652033090592,
|
|
"num_tokens": 8415220.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.207485580444336,
|
|
"epoch": 3.8158995815899583,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003925484165802201,
|
|
"loss": 4.907,
|
|
"mean_token_accuracy": 0.21459681540727615,
|
|
"num_tokens": 8424719.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.195284509658814,
|
|
"epoch": 3.8200836820083683,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003922730800617408,
|
|
"loss": 4.8177,
|
|
"mean_token_accuracy": 0.2186732456088066,
|
|
"num_tokens": 8433844.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.1423557758331295,
|
|
"epoch": 3.8242677824267783,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003919975022141174,
|
|
"loss": 4.7451,
|
|
"mean_token_accuracy": 0.22238616943359374,
|
|
"num_tokens": 8442700.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.310567712783813,
|
|
"epoch": 3.8284518828451883,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00039172168360444463,
|
|
"loss": 4.9802,
|
|
"mean_token_accuracy": 0.20577452927827836,
|
|
"num_tokens": 8452246.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.232901763916016,
|
|
"epoch": 3.8326359832635983,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039144562480031297,
|
|
"loss": 4.8338,
|
|
"mean_token_accuracy": 0.2197889968752861,
|
|
"num_tokens": 8461977.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.150258779525757,
|
|
"epoch": 3.8368200836820083,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039116932636980685,
|
|
"loss": 4.7387,
|
|
"mean_token_accuracy": 0.22440471351146699,
|
|
"num_tokens": 8470124.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.177676868438721,
|
|
"epoch": 3.8410041841004183,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039089278888150417,
|
|
"loss": 4.8617,
|
|
"mean_token_accuracy": 0.20720163732767105,
|
|
"num_tokens": 8479408.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.271122932434082,
|
|
"epoch": 3.8451882845188283,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003906160129044746,
|
|
"loss": 4.8121,
|
|
"mean_token_accuracy": 0.2202397108078003,
|
|
"num_tokens": 8488239.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.332623863220215,
|
|
"epoch": 3.8493723849372383,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003903389990082787,
|
|
"loss": 4.8628,
|
|
"mean_token_accuracy": 0.20828416645526887,
|
|
"num_tokens": 8497690.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.242746734619141,
|
|
"epoch": 3.8535564853556483,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003900617477629663,
|
|
"loss": 4.9366,
|
|
"mean_token_accuracy": 0.21114220172166825,
|
|
"num_tokens": 8506974.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.230209016799927,
|
|
"epoch": 3.8577405857740583,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003897842597390761,
|
|
"loss": 4.874,
|
|
"mean_token_accuracy": 0.21849111020565032,
|
|
"num_tokens": 8516066.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.246339988708496,
|
|
"epoch": 3.8619246861924688,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003895065355076338,
|
|
"loss": 4.8369,
|
|
"mean_token_accuracy": 0.21447764635086058,
|
|
"num_tokens": 8524747.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.177663993835449,
|
|
"epoch": 3.8661087866108788,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003892285756401511,
|
|
"loss": 4.821,
|
|
"mean_token_accuracy": 0.2167231023311615,
|
|
"num_tokens": 8534274.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.261530780792237,
|
|
"epoch": 3.8702928870292888,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003889503807086248,
|
|
"loss": 4.8346,
|
|
"mean_token_accuracy": 0.21081115454435348,
|
|
"num_tokens": 8543453.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.2339674949646,
|
|
"epoch": 3.8744769874476988,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038867195128553536,
|
|
"loss": 4.9112,
|
|
"mean_token_accuracy": 0.2116708680987358,
|
|
"num_tokens": 8553190.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.226614046096802,
|
|
"epoch": 3.8786610878661087,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003883932879438457,
|
|
"loss": 4.7951,
|
|
"mean_token_accuracy": 0.2100687101483345,
|
|
"num_tokens": 8562814.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.213352346420288,
|
|
"epoch": 3.8828451882845187,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003881143912570002,
|
|
"loss": 4.8065,
|
|
"mean_token_accuracy": 0.2139836549758911,
|
|
"num_tokens": 8572131.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.180496025085449,
|
|
"epoch": 3.8870292887029287,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003878352617989235,
|
|
"loss": 4.8877,
|
|
"mean_token_accuracy": 0.20415620058774947,
|
|
"num_tokens": 8582325.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.193806600570679,
|
|
"epoch": 3.891213389121339,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003875559001440191,
|
|
"loss": 4.7938,
|
|
"mean_token_accuracy": 0.2193414866924286,
|
|
"num_tokens": 8591704.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.2458336353302,
|
|
"epoch": 3.895397489539749,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003872763068671682,
|
|
"loss": 4.8467,
|
|
"mean_token_accuracy": 0.20663511604070664,
|
|
"num_tokens": 8601136.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.189525604248047,
|
|
"epoch": 3.899581589958159,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003869964825437291,
|
|
"loss": 4.8221,
|
|
"mean_token_accuracy": 0.21979267448186873,
|
|
"num_tokens": 8610491.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.269828510284424,
|
|
"epoch": 3.903765690376569,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00038671642774953507,
|
|
"loss": 4.8481,
|
|
"mean_token_accuracy": 0.21626490950584412,
|
|
"num_tokens": 8619199.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.251129579544068,
|
|
"epoch": 3.907949790794979,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003864361430608939,
|
|
"loss": 4.9284,
|
|
"mean_token_accuracy": 0.21175555884838104,
|
|
"num_tokens": 8629291.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.217710018157959,
|
|
"epoch": 3.912133891213389,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003861556290545864,
|
|
"loss": 4.8629,
|
|
"mean_token_accuracy": 0.21104538887739183,
|
|
"num_tokens": 8639212.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.207861089706421,
|
|
"epoch": 3.916317991631799,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038587488630786534,
|
|
"loss": 4.8233,
|
|
"mean_token_accuracy": 0.2131186917424202,
|
|
"num_tokens": 8648351.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.27808027267456,
|
|
"epoch": 3.920502092050209,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038559391539845413,
|
|
"loss": 4.8672,
|
|
"mean_token_accuracy": 0.20627767443656922,
|
|
"num_tokens": 8657267.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.241433477401733,
|
|
"epoch": 3.924686192468619,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038531271690454567,
|
|
"loss": 5.0101,
|
|
"mean_token_accuracy": 0.20444543957710265,
|
|
"num_tokens": 8666982.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.247953128814697,
|
|
"epoch": 3.928870292887029,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003850312914048015,
|
|
"loss": 4.8041,
|
|
"mean_token_accuracy": 0.22239669561386108,
|
|
"num_tokens": 8676067.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.302038669586182,
|
|
"epoch": 3.933054393305439,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003847496394783496,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.22180999517440797,
|
|
"num_tokens": 8685106.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.257042360305786,
|
|
"epoch": 3.937238493723849,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003844677617047848,
|
|
"loss": 4.9244,
|
|
"mean_token_accuracy": 0.20488815605640412,
|
|
"num_tokens": 8695203.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.090216970443725,
|
|
"epoch": 3.941422594142259,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000384185658664166,
|
|
"loss": 4.737,
|
|
"mean_token_accuracy": 0.21999370008707048,
|
|
"num_tokens": 8703700.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.254677724838257,
|
|
"epoch": 3.9456066945606696,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038390333093701596,
|
|
"loss": 4.8951,
|
|
"mean_token_accuracy": 0.20806924402713775,
|
|
"num_tokens": 8713393.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.251090621948242,
|
|
"epoch": 3.9497907949790796,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00038362077910431973,
|
|
"loss": 4.8334,
|
|
"mean_token_accuracy": 0.21570540517568587,
|
|
"num_tokens": 8722263.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.193048810958862,
|
|
"epoch": 3.9539748953974896,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003833380037475236,
|
|
"loss": 4.8354,
|
|
"mean_token_accuracy": 0.21554648876190186,
|
|
"num_tokens": 8732190.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.270418739318847,
|
|
"epoch": 3.9581589958158996,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00038305500544853375,
|
|
"loss": 4.8102,
|
|
"mean_token_accuracy": 0.22482340335845946,
|
|
"num_tokens": 8740995.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.29265832901001,
|
|
"epoch": 3.9623430962343096,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038277178478971527,
|
|
"loss": 4.9359,
|
|
"mean_token_accuracy": 0.206074495613575,
|
|
"num_tokens": 8750768.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.212096643447876,
|
|
"epoch": 3.9665271966527196,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003824883423538905,
|
|
"loss": 4.8412,
|
|
"mean_token_accuracy": 0.22013393491506578,
|
|
"num_tokens": 8760070.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.17306489944458,
|
|
"epoch": 3.9707112970711296,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038220467872433875,
|
|
"loss": 4.8415,
|
|
"mean_token_accuracy": 0.22107450813055038,
|
|
"num_tokens": 8769313.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.222470092773437,
|
|
"epoch": 3.97489539748954,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003819207944847939,
|
|
"loss": 4.8199,
|
|
"mean_token_accuracy": 0.216075199842453,
|
|
"num_tokens": 8778328.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.237242984771728,
|
|
"epoch": 3.97907949790795,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038163669021944413,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.21896411925554277,
|
|
"num_tokens": 8787408.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.232561683654785,
|
|
"epoch": 3.98326359832636,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003813523665129304,
|
|
"loss": 4.7916,
|
|
"mean_token_accuracy": 0.21729977428913116,
|
|
"num_tokens": 8796781.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.252694129943848,
|
|
"epoch": 3.98744769874477,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003810678239503452,
|
|
"loss": 4.8456,
|
|
"mean_token_accuracy": 0.21401668190956116,
|
|
"num_tokens": 8806965.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.1531929016113285,
|
|
"epoch": 3.99163179916318,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003807830631172312,
|
|
"loss": 4.8519,
|
|
"mean_token_accuracy": 0.21614866703748703,
|
|
"num_tokens": 8817348.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.207834243774414,
|
|
"epoch": 3.99581589958159,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038049808459958065,
|
|
"loss": 4.8939,
|
|
"mean_token_accuracy": 0.21536750942468644,
|
|
"num_tokens": 8827304.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.23601450920105,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038021288898383343,
|
|
"loss": 4.815,
|
|
"mean_token_accuracy": 0.2174364984035492,
|
|
"num_tokens": 8837288.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.25702748298645,
|
|
"epoch": 4.00418410041841,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003799274768568761,
|
|
"loss": 4.6324,
|
|
"mean_token_accuracy": 0.23598840236663818,
|
|
"num_tokens": 8846447.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.191226005554199,
|
|
"epoch": 4.00836820083682,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037964184880604125,
|
|
"loss": 4.5469,
|
|
"mean_token_accuracy": 0.2396013170480728,
|
|
"num_tokens": 8855990.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.109043836593628,
|
|
"epoch": 4.01255230125523,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00037935600541910523,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.23205065876245498,
|
|
"num_tokens": 8864949.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.111369705200195,
|
|
"epoch": 4.01673640167364,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003790699472842879,
|
|
"loss": 4.4979,
|
|
"mean_token_accuracy": 0.24599767476320267,
|
|
"num_tokens": 8873501.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.165746736526489,
|
|
"epoch": 4.02092050209205,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037878367499025086,
|
|
"loss": 4.609,
|
|
"mean_token_accuracy": 0.2320401668548584,
|
|
"num_tokens": 8882920.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.1607872486114506,
|
|
"epoch": 4.02510460251046,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003784971891260965,
|
|
"loss": 4.562,
|
|
"mean_token_accuracy": 0.23650027215480804,
|
|
"num_tokens": 8893617.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.189704275131225,
|
|
"epoch": 4.02928870292887,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037821049028136664,
|
|
"loss": 4.546,
|
|
"mean_token_accuracy": 0.23383972346782683,
|
|
"num_tokens": 8902558.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.16922287940979,
|
|
"epoch": 4.03347280334728,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003779235790460414,
|
|
"loss": 4.5796,
|
|
"mean_token_accuracy": 0.23885808289051055,
|
|
"num_tokens": 8912477.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.156034517288208,
|
|
"epoch": 4.03765690376569,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003776364560105379,
|
|
"loss": 4.5964,
|
|
"mean_token_accuracy": 0.23315861970186233,
|
|
"num_tokens": 8922313.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.168852043151856,
|
|
"epoch": 4.0418410041841,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00037734912176570935,
|
|
"loss": 4.5405,
|
|
"mean_token_accuracy": 0.24275935292243958,
|
|
"num_tokens": 8932011.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.153744411468506,
|
|
"epoch": 4.046025104602511,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037706157690284335,
|
|
"loss": 4.5949,
|
|
"mean_token_accuracy": 0.23160350173711777,
|
|
"num_tokens": 8941022.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.169446754455566,
|
|
"epoch": 4.050209205020921,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037677382201366103,
|
|
"loss": 4.6071,
|
|
"mean_token_accuracy": 0.24071440994739532,
|
|
"num_tokens": 8950825.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.1695881366729735,
|
|
"epoch": 4.054393305439331,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037648585769031555,
|
|
"loss": 4.6224,
|
|
"mean_token_accuracy": 0.22795521169900895,
|
|
"num_tokens": 8961035.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.2240081310272215,
|
|
"epoch": 4.058577405857741,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003761976845253915,
|
|
"loss": 4.654,
|
|
"mean_token_accuracy": 0.2337782472372055,
|
|
"num_tokens": 8970535.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.1366307735443115,
|
|
"epoch": 4.062761506276151,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003759093031119025,
|
|
"loss": 4.5743,
|
|
"mean_token_accuracy": 0.23261162489652634,
|
|
"num_tokens": 8979404.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.203960704803467,
|
|
"epoch": 4.066945606694561,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037562071404329147,
|
|
"loss": 4.6121,
|
|
"mean_token_accuracy": 0.24727526009082795,
|
|
"num_tokens": 8989164.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.2392627716064455,
|
|
"epoch": 4.071129707112971,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00037533191791342817,
|
|
"loss": 4.6311,
|
|
"mean_token_accuracy": 0.22781284898519516,
|
|
"num_tokens": 8998630.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.178066778182983,
|
|
"epoch": 4.075313807531381,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037504291531660866,
|
|
"loss": 4.5642,
|
|
"mean_token_accuracy": 0.24225292205810547,
|
|
"num_tokens": 9008235.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.185148239135742,
|
|
"epoch": 4.079497907949791,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003747537068475538,
|
|
"loss": 4.6788,
|
|
"mean_token_accuracy": 0.23501088768243789,
|
|
"num_tokens": 9018175.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.267748403549194,
|
|
"epoch": 4.083682008368201,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003744642931014082,
|
|
"loss": 4.6956,
|
|
"mean_token_accuracy": 0.23158361315727233,
|
|
"num_tokens": 9028159.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.103841352462768,
|
|
"epoch": 4.087866108786611,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003741746746737386,
|
|
"loss": 4.603,
|
|
"mean_token_accuracy": 0.23099596053361893,
|
|
"num_tokens": 9038263.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.155388355255127,
|
|
"epoch": 4.092050209205021,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003738848521605333,
|
|
"loss": 4.5857,
|
|
"mean_token_accuracy": 0.2384982869029045,
|
|
"num_tokens": 9048023.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.1613850593566895,
|
|
"epoch": 4.096234309623431,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003735948261582005,
|
|
"loss": 4.6575,
|
|
"mean_token_accuracy": 0.22373982816934584,
|
|
"num_tokens": 9056988.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.171466541290283,
|
|
"epoch": 4.100418410041841,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000373304597263567,
|
|
"loss": 4.6138,
|
|
"mean_token_accuracy": 0.2348845988512039,
|
|
"num_tokens": 9066595.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.179251956939697,
|
|
"epoch": 4.104602510460251,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00037301416607387716,
|
|
"loss": 4.6421,
|
|
"mean_token_accuracy": 0.23143157809972764,
|
|
"num_tokens": 9075034.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.113850545883179,
|
|
"epoch": 4.108786610878661,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037272353318679185,
|
|
"loss": 4.5806,
|
|
"mean_token_accuracy": 0.2391144812107086,
|
|
"num_tokens": 9084574.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.105892133712769,
|
|
"epoch": 4.112970711297071,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003724326992003867,
|
|
"loss": 4.5558,
|
|
"mean_token_accuracy": 0.23474373370409013,
|
|
"num_tokens": 9093591.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.1952625751495365,
|
|
"epoch": 4.117154811715481,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003721416647131512,
|
|
"loss": 4.615,
|
|
"mean_token_accuracy": 0.23617635816335678,
|
|
"num_tokens": 9102439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.119270658493042,
|
|
"epoch": 4.121338912133891,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003718504303239876,
|
|
"loss": 4.4677,
|
|
"mean_token_accuracy": 0.24569525122642516,
|
|
"num_tokens": 9111363.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.182420873641968,
|
|
"epoch": 4.125523012552302,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003715589966322097,
|
|
"loss": 4.664,
|
|
"mean_token_accuracy": 0.2345169886946678,
|
|
"num_tokens": 9120417.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.1268613815307615,
|
|
"epoch": 4.129707112970712,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003712673642375409,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.22753059566020967,
|
|
"num_tokens": 9130622.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.113135766983032,
|
|
"epoch": 4.133891213389122,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003709755337401138,
|
|
"loss": 4.6208,
|
|
"mean_token_accuracy": 0.231180639564991,
|
|
"num_tokens": 9139992.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.121628618240356,
|
|
"epoch": 4.138075313807532,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037068350574046867,
|
|
"loss": 4.5768,
|
|
"mean_token_accuracy": 0.23461578786373138,
|
|
"num_tokens": 9149883.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.110263156890869,
|
|
"epoch": 4.142259414225942,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037039128083955227,
|
|
"loss": 4.564,
|
|
"mean_token_accuracy": 0.2437523826956749,
|
|
"num_tokens": 9159633.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.129648160934448,
|
|
"epoch": 4.146443514644352,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003700988596387163,
|
|
"loss": 4.5506,
|
|
"mean_token_accuracy": 0.24082456678152084,
|
|
"num_tokens": 9167780.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.144667243957519,
|
|
"epoch": 4.150627615062762,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003698062427397168,
|
|
"loss": 4.5798,
|
|
"mean_token_accuracy": 0.23522408306598663,
|
|
"num_tokens": 9176808.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.193084239959717,
|
|
"epoch": 4.154811715481172,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036951343074471217,
|
|
"loss": 4.6042,
|
|
"mean_token_accuracy": 0.23875458538532257,
|
|
"num_tokens": 9186097.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.156343221664429,
|
|
"epoch": 4.158995815899582,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003692204242562624,
|
|
"loss": 4.6379,
|
|
"mean_token_accuracy": 0.22362160682678223,
|
|
"num_tokens": 9194428.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.136697101593017,
|
|
"epoch": 4.163179916317992,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003689272238773278,
|
|
"loss": 4.6275,
|
|
"mean_token_accuracy": 0.2332097202539444,
|
|
"num_tokens": 9203156.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.12443437576294,
|
|
"epoch": 4.167364016736402,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00036863383021126756,
|
|
"loss": 4.5969,
|
|
"mean_token_accuracy": 0.2378309190273285,
|
|
"num_tokens": 9212102.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.138901615142823,
|
|
"epoch": 4.171548117154812,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036834024386183875,
|
|
"loss": 4.6269,
|
|
"mean_token_accuracy": 0.230842524766922,
|
|
"num_tokens": 9221507.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.132518339157104,
|
|
"epoch": 4.175732217573222,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003680464654331947,
|
|
"loss": 4.5633,
|
|
"mean_token_accuracy": 0.2392383337020874,
|
|
"num_tokens": 9230543.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.214751672744751,
|
|
"epoch": 4.179916317991632,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003677524955298844,
|
|
"loss": 4.6749,
|
|
"mean_token_accuracy": 0.22901857793331146,
|
|
"num_tokens": 9240092.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.137883424758911,
|
|
"epoch": 4.184100418410042,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003674583347568505,
|
|
"loss": 4.5917,
|
|
"mean_token_accuracy": 0.23220259249210357,
|
|
"num_tokens": 9250145.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.184100418410042,
|
|
"eval_entropy": 4.911186488371929,
|
|
"eval_loss": 5.316469192504883,
|
|
"eval_mean_token_accuracy": 0.19883140129301932,
|
|
"eval_num_tokens": 9250145.0,
|
|
"eval_runtime": 2.0932,
|
|
"eval_samples_per_second": 1866.479,
|
|
"eval_steps_per_second": 233.608,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"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": 2001685689630720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|