e9545c4dde
Model: fpadovani/dan-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.41841004184100417,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 200351734824960.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|