7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.1518386714116255,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.6509840965271,
|
|
"epoch": 0.005931198102016607,
|
|
"grad_norm": 11.4375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7955,
|
|
"mean_token_accuracy": 0.00017467249417677522,
|
|
"num_tokens": 12647.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.650976181030273,
|
|
"epoch": 0.011862396204033215,
|
|
"grad_norm": 10.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7449,
|
|
"mean_token_accuracy": 0.0003256943717133254,
|
|
"num_tokens": 26668.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.65047550201416,
|
|
"epoch": 0.017793594306049824,
|
|
"grad_norm": 8.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5392,
|
|
"mean_token_accuracy": 0.011320227157557382,
|
|
"num_tokens": 41021.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.645245170593261,
|
|
"epoch": 0.02372479240806643,
|
|
"grad_norm": 5.6875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2512,
|
|
"mean_token_accuracy": 0.03331172131001949,
|
|
"num_tokens": 56626.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.620162200927734,
|
|
"epoch": 0.02965599051008304,
|
|
"grad_norm": 3.46875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9999,
|
|
"mean_token_accuracy": 0.03278177715837956,
|
|
"num_tokens": 72914.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.59833517074585,
|
|
"epoch": 0.03558718861209965,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8507,
|
|
"mean_token_accuracy": 0.03321463130414486,
|
|
"num_tokens": 88514.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.59533166885376,
|
|
"epoch": 0.04151838671411625,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7557,
|
|
"mean_token_accuracy": 0.03128496501594782,
|
|
"num_tokens": 101256.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.598009204864502,
|
|
"epoch": 0.04744958481613286,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7229,
|
|
"mean_token_accuracy": 0.03391152955591679,
|
|
"num_tokens": 114459.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.583034133911132,
|
|
"epoch": 0.05338078291814947,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.7096,
|
|
"mean_token_accuracy": 0.031900213845074174,
|
|
"num_tokens": 129858.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.583144092559815,
|
|
"epoch": 0.05931198102016608,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6353,
|
|
"mean_token_accuracy": 0.03371269796043634,
|
|
"num_tokens": 143486.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.582204055786132,
|
|
"epoch": 0.06524317912218268,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5795,
|
|
"mean_token_accuracy": 0.035220497474074366,
|
|
"num_tokens": 156003.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.569349002838134,
|
|
"epoch": 0.0711743772241993,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5437,
|
|
"mean_token_accuracy": 0.034727090038359165,
|
|
"num_tokens": 170988.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.552780246734619,
|
|
"epoch": 0.0771055753262159,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4193,
|
|
"mean_token_accuracy": 0.03448135014623403,
|
|
"num_tokens": 187165.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.542837047576905,
|
|
"epoch": 0.0830367734282325,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.368,
|
|
"mean_token_accuracy": 0.03228826243430376,
|
|
"num_tokens": 202072.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.534712505340575,
|
|
"epoch": 0.08896797153024912,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2131,
|
|
"mean_token_accuracy": 0.034324941039085385,
|
|
"num_tokens": 215375.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.50705509185791,
|
|
"epoch": 0.09489916963226572,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1257,
|
|
"mean_token_accuracy": 0.03353528436273336,
|
|
"num_tokens": 228749.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.478670978546143,
|
|
"epoch": 0.10083036773428232,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0183,
|
|
"mean_token_accuracy": 0.03314766474068165,
|
|
"num_tokens": 242387.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.458618545532227,
|
|
"epoch": 0.10676156583629894,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9252,
|
|
"mean_token_accuracy": 0.03416665960103273,
|
|
"num_tokens": 257793.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.419714641571044,
|
|
"epoch": 0.11269276393831554,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8134,
|
|
"mean_token_accuracy": 0.036843133717775346,
|
|
"num_tokens": 271187.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.339355945587158,
|
|
"epoch": 0.11862396204033215,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.6836,
|
|
"mean_token_accuracy": 0.04052316285669803,
|
|
"num_tokens": 284937.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.289305877685546,
|
|
"epoch": 0.12455516014234876,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.523,
|
|
"mean_token_accuracy": 0.042971277609467506,
|
|
"num_tokens": 300108.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.171844291687012,
|
|
"epoch": 0.13048635824436536,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4071,
|
|
"mean_token_accuracy": 0.0435908630490303,
|
|
"num_tokens": 316760.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.090029907226562,
|
|
"epoch": 0.13641755634638197,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2269,
|
|
"mean_token_accuracy": 0.04757884480059147,
|
|
"num_tokens": 329705.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.968478775024414,
|
|
"epoch": 0.1423487544483986,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.1048,
|
|
"mean_token_accuracy": 0.05177154988050461,
|
|
"num_tokens": 342947.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.810734272003174,
|
|
"epoch": 0.14827995255041518,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.957,
|
|
"mean_token_accuracy": 0.05589810460805893,
|
|
"num_tokens": 358397.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.629272270202637,
|
|
"epoch": 0.1542111506524318,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8543,
|
|
"mean_token_accuracy": 0.058608030527830125,
|
|
"num_tokens": 370369.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.431285285949707,
|
|
"epoch": 0.1601423487544484,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7282,
|
|
"mean_token_accuracy": 0.06019565127789974,
|
|
"num_tokens": 386896.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.166491031646729,
|
|
"epoch": 0.166073546856465,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.571,
|
|
"mean_token_accuracy": 0.06539936549961567,
|
|
"num_tokens": 400433.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.998315811157227,
|
|
"epoch": 0.17200474495848161,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5193,
|
|
"mean_token_accuracy": 0.06127564944326878,
|
|
"num_tokens": 415092.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.760649585723877,
|
|
"epoch": 0.17793594306049823,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.3847,
|
|
"mean_token_accuracy": 0.06551449336111545,
|
|
"num_tokens": 429793.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.53032569885254,
|
|
"epoch": 0.18386714116251482,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3618,
|
|
"mean_token_accuracy": 0.07152369543910027,
|
|
"num_tokens": 442618.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.432501602172852,
|
|
"epoch": 0.18979833926453143,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.2263,
|
|
"mean_token_accuracy": 0.0709031268954277,
|
|
"num_tokens": 454976.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.288888359069825,
|
|
"epoch": 0.19572953736654805,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2693,
|
|
"mean_token_accuracy": 0.06482412219047547,
|
|
"num_tokens": 467523.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.135635375976562,
|
|
"epoch": 0.20166073546856464,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.1729,
|
|
"mean_token_accuracy": 0.07200234681367874,
|
|
"num_tokens": 481391.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.090270137786865,
|
|
"epoch": 0.20759193357058126,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.147,
|
|
"mean_token_accuracy": 0.06691372841596603,
|
|
"num_tokens": 496697.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.008182764053345,
|
|
"epoch": 0.21352313167259787,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1704,
|
|
"mean_token_accuracy": 0.07276545092463493,
|
|
"num_tokens": 510091.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.948391151428223,
|
|
"epoch": 0.21945432977461446,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.1071,
|
|
"mean_token_accuracy": 0.0736262746155262,
|
|
"num_tokens": 522168.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.860538482666016,
|
|
"epoch": 0.22538552787663108,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.0333,
|
|
"mean_token_accuracy": 0.0750058189034462,
|
|
"num_tokens": 535853.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.789449167251587,
|
|
"epoch": 0.2313167259786477,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.0532,
|
|
"mean_token_accuracy": 0.07566255033016205,
|
|
"num_tokens": 551082.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.820679521560669,
|
|
"epoch": 0.2372479240806643,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.0735,
|
|
"mean_token_accuracy": 0.0763503935188055,
|
|
"num_tokens": 567181.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.715967988967895,
|
|
"epoch": 0.2431791221826809,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.9882,
|
|
"mean_token_accuracy": 0.0811008334159851,
|
|
"num_tokens": 580689.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.671104192733765,
|
|
"epoch": 0.2491103202846975,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.9075,
|
|
"mean_token_accuracy": 0.08624595254659653,
|
|
"num_tokens": 594590.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.669063520431519,
|
|
"epoch": 0.25504151838671413,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.9659,
|
|
"mean_token_accuracy": 0.07704100087285042,
|
|
"num_tokens": 610577.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.621254777908325,
|
|
"epoch": 0.2609727164887307,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.9894,
|
|
"mean_token_accuracy": 0.0770975686609745,
|
|
"num_tokens": 627066.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.582563304901123,
|
|
"epoch": 0.2669039145907473,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.9863,
|
|
"mean_token_accuracy": 0.08030139058828353,
|
|
"num_tokens": 642601.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.5748522758483885,
|
|
"epoch": 0.27283511269276395,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.8436,
|
|
"mean_token_accuracy": 0.08795318827033043,
|
|
"num_tokens": 655615.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.51125283241272,
|
|
"epoch": 0.27876631079478054,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.8805,
|
|
"mean_token_accuracy": 0.08227636963129044,
|
|
"num_tokens": 670218.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.581772565841675,
|
|
"epoch": 0.2846975088967972,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.8693,
|
|
"mean_token_accuracy": 0.08509634733200074,
|
|
"num_tokens": 683661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.414237833023071,
|
|
"epoch": 0.29062870699881377,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.8471,
|
|
"mean_token_accuracy": 0.08358772993087768,
|
|
"num_tokens": 699767.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.437202262878418,
|
|
"epoch": 0.29655990510083036,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.774,
|
|
"mean_token_accuracy": 0.09326352775096894,
|
|
"num_tokens": 713579.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.383161449432373,
|
|
"epoch": 0.302491103202847,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.8019,
|
|
"mean_token_accuracy": 0.08484338372945785,
|
|
"num_tokens": 727490.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.414105653762817,
|
|
"epoch": 0.3084223013048636,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.8822,
|
|
"mean_token_accuracy": 0.08486307561397552,
|
|
"num_tokens": 742477.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.386582565307617,
|
|
"epoch": 0.3143534994068802,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.7436,
|
|
"mean_token_accuracy": 0.08875370100140571,
|
|
"num_tokens": 757772.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.385796451568604,
|
|
"epoch": 0.3202846975088968,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.795,
|
|
"mean_token_accuracy": 0.08812053054571152,
|
|
"num_tokens": 772898.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.329386568069458,
|
|
"epoch": 0.3262158956109134,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.7697,
|
|
"mean_token_accuracy": 0.08715710267424584,
|
|
"num_tokens": 788395.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.318710279464722,
|
|
"epoch": 0.33214709371293,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.83,
|
|
"mean_token_accuracy": 0.08070869520306587,
|
|
"num_tokens": 803413.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.188418292999268,
|
|
"epoch": 0.33807829181494664,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.6976,
|
|
"mean_token_accuracy": 0.09588895291090012,
|
|
"num_tokens": 817775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.389742612838745,
|
|
"epoch": 0.34400948991696323,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8092,
|
|
"mean_token_accuracy": 0.08903076201677322,
|
|
"num_tokens": 831379.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.300806474685669,
|
|
"epoch": 0.3499406880189798,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.7387,
|
|
"mean_token_accuracy": 0.08791814595460892,
|
|
"num_tokens": 846596.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.254555749893188,
|
|
"epoch": 0.35587188612099646,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.7438,
|
|
"mean_token_accuracy": 0.08821789696812629,
|
|
"num_tokens": 862063.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.275745964050293,
|
|
"epoch": 0.36180308422301305,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.6954,
|
|
"mean_token_accuracy": 0.08966124951839446,
|
|
"num_tokens": 876841.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.221111488342285,
|
|
"epoch": 0.36773428232502964,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.7751,
|
|
"mean_token_accuracy": 0.09116524010896683,
|
|
"num_tokens": 890728.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.206338787078858,
|
|
"epoch": 0.3736654804270463,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.6956,
|
|
"mean_token_accuracy": 0.09070762842893601,
|
|
"num_tokens": 905599.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.1817542552948,
|
|
"epoch": 0.37959667852906287,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.6483,
|
|
"mean_token_accuracy": 0.09768397733569145,
|
|
"num_tokens": 918139.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.167359304428101,
|
|
"epoch": 0.38552787663107946,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.7114,
|
|
"mean_token_accuracy": 0.09328882694244385,
|
|
"num_tokens": 931666.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.196693468093872,
|
|
"epoch": 0.3914590747330961,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.09237670227885246,
|
|
"num_tokens": 946286.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.148892021179199,
|
|
"epoch": 0.3973902728351127,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.09802859723567962,
|
|
"num_tokens": 959689.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.091142225265503,
|
|
"epoch": 0.4033214709371293,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.6481,
|
|
"mean_token_accuracy": 0.09793345853686333,
|
|
"num_tokens": 972707.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.103200006484985,
|
|
"epoch": 0.4092526690391459,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.6694,
|
|
"mean_token_accuracy": 0.09691287726163864,
|
|
"num_tokens": 988010.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.109372472763061,
|
|
"epoch": 0.4151838671411625,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.7001,
|
|
"mean_token_accuracy": 0.0866947390139103,
|
|
"num_tokens": 1003823.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.018935489654541,
|
|
"epoch": 0.4211150652431791,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.4583,
|
|
"mean_token_accuracy": 0.10792901739478111,
|
|
"num_tokens": 1018276.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.151296997070313,
|
|
"epoch": 0.42704626334519574,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.6782,
|
|
"mean_token_accuracy": 0.09361589401960373,
|
|
"num_tokens": 1031242.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.0902214527130125,
|
|
"epoch": 0.43297746144721233,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.5786,
|
|
"mean_token_accuracy": 0.09998262599110604,
|
|
"num_tokens": 1043940.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.944676065444947,
|
|
"epoch": 0.4389086595492289,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.508,
|
|
"mean_token_accuracy": 0.09759898036718369,
|
|
"num_tokens": 1058656.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.117189264297485,
|
|
"epoch": 0.44483985765124556,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6092,
|
|
"mean_token_accuracy": 0.09189791083335877,
|
|
"num_tokens": 1073128.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.948076343536377,
|
|
"epoch": 0.45077105575326215,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.5416,
|
|
"mean_token_accuracy": 0.09809970185160637,
|
|
"num_tokens": 1086467.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.076186227798462,
|
|
"epoch": 0.45670225385527874,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6616,
|
|
"mean_token_accuracy": 0.10060962289571762,
|
|
"num_tokens": 1100356.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.028731060028076,
|
|
"epoch": 0.4626334519572954,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.5814,
|
|
"mean_token_accuracy": 0.10232764557003975,
|
|
"num_tokens": 1115877.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.04663872718811,
|
|
"epoch": 0.46856465005931197,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.5466,
|
|
"mean_token_accuracy": 0.10219399407505989,
|
|
"num_tokens": 1131415.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.8535503387451175,
|
|
"epoch": 0.4744958481613286,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.5137,
|
|
"mean_token_accuracy": 0.09734094068408013,
|
|
"num_tokens": 1145903.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.879596471786499,
|
|
"epoch": 0.4804270462633452,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.0997365266084671,
|
|
"num_tokens": 1161168.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.921687078475952,
|
|
"epoch": 0.4863582443653618,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5196,
|
|
"mean_token_accuracy": 0.09855128526687622,
|
|
"num_tokens": 1178155.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.924295806884766,
|
|
"epoch": 0.49228944246737844,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.3697,
|
|
"mean_token_accuracy": 0.1099881686270237,
|
|
"num_tokens": 1190830.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.827713060379028,
|
|
"epoch": 0.498220640569395,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.3842,
|
|
"mean_token_accuracy": 0.10342236086726189,
|
|
"num_tokens": 1205260.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.883217096328735,
|
|
"epoch": 0.5041518386714117,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.431,
|
|
"mean_token_accuracy": 0.10836012288928032,
|
|
"num_tokens": 1218856.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.832102727890015,
|
|
"epoch": 0.5100830367734283,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5014,
|
|
"mean_token_accuracy": 0.10428635627031327,
|
|
"num_tokens": 1234669.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.830685758590699,
|
|
"epoch": 0.5160142348754448,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.4289,
|
|
"mean_token_accuracy": 0.1057804524898529,
|
|
"num_tokens": 1249514.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.837182474136353,
|
|
"epoch": 0.5219454329774614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.3638,
|
|
"mean_token_accuracy": 0.1109524667263031,
|
|
"num_tokens": 1262311.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.767211008071899,
|
|
"epoch": 0.527876631079478,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.3519,
|
|
"mean_token_accuracy": 0.10807539224624634,
|
|
"num_tokens": 1277681.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.781484460830688,
|
|
"epoch": 0.5338078291814946,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.3968,
|
|
"mean_token_accuracy": 0.10917404815554618,
|
|
"num_tokens": 1292300.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.775958061218262,
|
|
"epoch": 0.5397390272835113,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.10393262058496475,
|
|
"num_tokens": 1307426.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.787585258483887,
|
|
"epoch": 0.5456702253855279,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.4414,
|
|
"mean_token_accuracy": 0.09955934658646584,
|
|
"num_tokens": 1322167.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.75556001663208,
|
|
"epoch": 0.5516014234875445,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.3126,
|
|
"mean_token_accuracy": 0.11029193922877312,
|
|
"num_tokens": 1335343.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.719364500045776,
|
|
"epoch": 0.5575326215895611,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.3129,
|
|
"mean_token_accuracy": 0.11020243093371392,
|
|
"num_tokens": 1349078.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.670861291885376,
|
|
"epoch": 0.5634638196915777,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.2388,
|
|
"mean_token_accuracy": 0.11813571006059646,
|
|
"num_tokens": 1362030.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.790364742279053,
|
|
"epoch": 0.5693950177935944,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.3965,
|
|
"mean_token_accuracy": 0.10456062778830529,
|
|
"num_tokens": 1378303.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.773904657363891,
|
|
"epoch": 0.575326215895611,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.3906,
|
|
"mean_token_accuracy": 0.10592009201645851,
|
|
"num_tokens": 1392585.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.805166101455688,
|
|
"epoch": 0.5812574139976275,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.3973,
|
|
"mean_token_accuracy": 0.11140659600496292,
|
|
"num_tokens": 1406962.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.573492240905762,
|
|
"epoch": 0.5871886120996441,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.2853,
|
|
"mean_token_accuracy": 0.11129999980330467,
|
|
"num_tokens": 1419493.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.735859537124634,
|
|
"epoch": 0.5931198102016607,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.3007,
|
|
"mean_token_accuracy": 0.10790175497531891,
|
|
"num_tokens": 1435373.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5931198102016607,
|
|
"eval_entropy": 6.67236985896994,
|
|
"eval_loss": 6.441715717315674,
|
|
"eval_mean_token_accuracy": 0.10966419813685988,
|
|
"eval_num_tokens": 1435373.0,
|
|
"eval_runtime": 2.4776,
|
|
"eval_samples_per_second": 1052.652,
|
|
"eval_steps_per_second": 131.581,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.6947290897369385,
|
|
"epoch": 0.5990510083036773,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.2707,
|
|
"mean_token_accuracy": 0.11756636276841163,
|
|
"num_tokens": 1449805.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.634400224685669,
|
|
"epoch": 0.604982206405694,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.2855,
|
|
"mean_token_accuracy": 0.11045260801911354,
|
|
"num_tokens": 1463657.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.727581596374511,
|
|
"epoch": 0.6109134045077106,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.11670920848846436,
|
|
"num_tokens": 1478089.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.619698333740234,
|
|
"epoch": 0.6168446026097272,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.311,
|
|
"mean_token_accuracy": 0.11146906316280365,
|
|
"num_tokens": 1492225.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.630937004089356,
|
|
"epoch": 0.6227758007117438,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.3008,
|
|
"mean_token_accuracy": 0.10728157833218574,
|
|
"num_tokens": 1506751.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.599562358856201,
|
|
"epoch": 0.6287069988137604,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.2004,
|
|
"mean_token_accuracy": 0.11617306619882584,
|
|
"num_tokens": 1520590.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.652629327774048,
|
|
"epoch": 0.6346381969157769,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.2527,
|
|
"mean_token_accuracy": 0.11343958005309104,
|
|
"num_tokens": 1534752.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.587781953811645,
|
|
"epoch": 0.6405693950177936,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.2154,
|
|
"mean_token_accuracy": 0.12041136622428894,
|
|
"num_tokens": 1547891.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.418690204620361,
|
|
"epoch": 0.6465005931198102,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.1542,
|
|
"mean_token_accuracy": 0.12026057541370391,
|
|
"num_tokens": 1562560.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.670684385299682,
|
|
"epoch": 0.6524317912218268,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.3091,
|
|
"mean_token_accuracy": 0.11120221018791199,
|
|
"num_tokens": 1576896.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.535441923141479,
|
|
"epoch": 0.6583629893238434,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.2086,
|
|
"mean_token_accuracy": 0.11225326210260392,
|
|
"num_tokens": 1590806.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.601208257675171,
|
|
"epoch": 0.66429418742586,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.326,
|
|
"mean_token_accuracy": 0.10773437917232513,
|
|
"num_tokens": 1605874.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.616091251373291,
|
|
"epoch": 0.6702253855278766,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.2108,
|
|
"mean_token_accuracy": 0.11664134711027145,
|
|
"num_tokens": 1620947.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.484318447113037,
|
|
"epoch": 0.6761565836298933,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.2608,
|
|
"mean_token_accuracy": 0.10255342498421668,
|
|
"num_tokens": 1635974.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.542778062820434,
|
|
"epoch": 0.6820877817319099,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.2163,
|
|
"mean_token_accuracy": 0.10959842428565025,
|
|
"num_tokens": 1651502.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.45567512512207,
|
|
"epoch": 0.6880189798339265,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.1629,
|
|
"mean_token_accuracy": 0.11728360131382942,
|
|
"num_tokens": 1665547.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.501108646392822,
|
|
"epoch": 0.693950177935943,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.1508,
|
|
"mean_token_accuracy": 0.12227020487189293,
|
|
"num_tokens": 1678875.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.480410623550415,
|
|
"epoch": 0.6998813760379596,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.1681,
|
|
"mean_token_accuracy": 0.11540260091423989,
|
|
"num_tokens": 1692673.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.511543416976929,
|
|
"epoch": 0.7058125741399762,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.2156,
|
|
"mean_token_accuracy": 0.11393610835075378,
|
|
"num_tokens": 1707318.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.460053396224976,
|
|
"epoch": 0.7117437722419929,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.2426,
|
|
"mean_token_accuracy": 0.11120611354708672,
|
|
"num_tokens": 1724331.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.442718982696533,
|
|
"epoch": 0.7176749703440095,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.0492,
|
|
"mean_token_accuracy": 0.12235584780573845,
|
|
"num_tokens": 1736151.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.444410896301269,
|
|
"epoch": 0.7236061684460261,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.1371,
|
|
"mean_token_accuracy": 0.11987931281328201,
|
|
"num_tokens": 1749831.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.455282926559448,
|
|
"epoch": 0.7295373665480427,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.173,
|
|
"mean_token_accuracy": 0.11412288546562195,
|
|
"num_tokens": 1764296.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.49344654083252,
|
|
"epoch": 0.7354685646500593,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.1473,
|
|
"mean_token_accuracy": 0.11842510327696801,
|
|
"num_tokens": 1778042.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.402311325073242,
|
|
"epoch": 0.741399762752076,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.1244,
|
|
"mean_token_accuracy": 0.1204920805990696,
|
|
"num_tokens": 1791614.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.386654949188232,
|
|
"epoch": 0.7473309608540926,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.082,
|
|
"mean_token_accuracy": 0.1170621544122696,
|
|
"num_tokens": 1807388.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.513570547103882,
|
|
"epoch": 0.7532621589561092,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.187,
|
|
"mean_token_accuracy": 0.1129879891872406,
|
|
"num_tokens": 1820407.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.353465366363525,
|
|
"epoch": 0.7591933570581257,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.11846939399838448,
|
|
"num_tokens": 1834666.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.402153253555298,
|
|
"epoch": 0.7651245551601423,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.169,
|
|
"mean_token_accuracy": 0.11969529762864113,
|
|
"num_tokens": 1848160.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.429388856887817,
|
|
"epoch": 0.7710557532621589,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.1763,
|
|
"mean_token_accuracy": 0.11832700744271278,
|
|
"num_tokens": 1865140.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.384272193908691,
|
|
"epoch": 0.7769869513641756,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.0502,
|
|
"mean_token_accuracy": 0.12135105207562447,
|
|
"num_tokens": 1878447.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.368830823898316,
|
|
"epoch": 0.7829181494661922,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.0999,
|
|
"mean_token_accuracy": 0.11421585902571678,
|
|
"num_tokens": 1893539.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.452864456176758,
|
|
"epoch": 0.7888493475682088,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.1646,
|
|
"mean_token_accuracy": 0.120599914342165,
|
|
"num_tokens": 1908052.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.347026681900024,
|
|
"epoch": 0.7947805456702254,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.0494,
|
|
"mean_token_accuracy": 0.12437219470739365,
|
|
"num_tokens": 1921528.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.360257863998413,
|
|
"epoch": 0.800711743772242,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.0758,
|
|
"mean_token_accuracy": 0.12383257448673249,
|
|
"num_tokens": 1935089.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.4030664443969725,
|
|
"epoch": 0.8066429418742586,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.1896,
|
|
"mean_token_accuracy": 0.11006285175681114,
|
|
"num_tokens": 1949717.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.300024366378784,
|
|
"epoch": 0.8125741399762753,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.0568,
|
|
"mean_token_accuracy": 0.12153587713837624,
|
|
"num_tokens": 1962619.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.450096464157104,
|
|
"epoch": 0.8185053380782918,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.143,
|
|
"mean_token_accuracy": 0.11284489929676056,
|
|
"num_tokens": 1977094.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.286441040039063,
|
|
"epoch": 0.8244365361803084,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.1106,
|
|
"mean_token_accuracy": 0.11766432374715804,
|
|
"num_tokens": 1992446.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.301593971252442,
|
|
"epoch": 0.830367734282325,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.0526,
|
|
"mean_token_accuracy": 0.1210049532353878,
|
|
"num_tokens": 2008753.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.312734651565552,
|
|
"epoch": 0.8362989323843416,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.0098,
|
|
"mean_token_accuracy": 0.1238975889980793,
|
|
"num_tokens": 2024088.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.338296222686767,
|
|
"epoch": 0.8422301304863582,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.0283,
|
|
"mean_token_accuracy": 0.11651979684829712,
|
|
"num_tokens": 2038449.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.306093740463257,
|
|
"epoch": 0.8481613285883749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.0258,
|
|
"mean_token_accuracy": 0.1278011307120323,
|
|
"num_tokens": 2052038.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.271956300735473,
|
|
"epoch": 0.8540925266903915,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 5.9769,
|
|
"mean_token_accuracy": 0.12276132255792618,
|
|
"num_tokens": 2064907.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.297043323516846,
|
|
"epoch": 0.8600237247924081,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.1502,
|
|
"mean_token_accuracy": 0.1173181675374508,
|
|
"num_tokens": 2081295.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.336274099349976,
|
|
"epoch": 0.8659549228944247,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.0652,
|
|
"mean_token_accuracy": 0.12099355086684227,
|
|
"num_tokens": 2098077.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.285319423675537,
|
|
"epoch": 0.8718861209964412,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.0532,
|
|
"mean_token_accuracy": 0.12013457342982292,
|
|
"num_tokens": 2111716.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.27389874458313,
|
|
"epoch": 0.8778173190984578,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 5.9654,
|
|
"mean_token_accuracy": 0.13291287645697594,
|
|
"num_tokens": 2125757.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.267977666854859,
|
|
"epoch": 0.8837485172004745,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.1618,
|
|
"mean_token_accuracy": 0.11648618057370186,
|
|
"num_tokens": 2142929.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.27850079536438,
|
|
"epoch": 0.8896797153024911,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.0379,
|
|
"mean_token_accuracy": 0.12373944073915481,
|
|
"num_tokens": 2155206.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.240651369094849,
|
|
"epoch": 0.8956109134045077,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 5.9246,
|
|
"mean_token_accuracy": 0.12547165155410767,
|
|
"num_tokens": 2169816.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.190544652938843,
|
|
"epoch": 0.9015421115065243,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.0044,
|
|
"mean_token_accuracy": 0.12672639563679694,
|
|
"num_tokens": 2184827.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.259784412384033,
|
|
"epoch": 0.9074733096085409,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.12988706678152084,
|
|
"num_tokens": 2199333.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.1021686553955075,
|
|
"epoch": 0.9134045077105575,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 5.9947,
|
|
"mean_token_accuracy": 0.12381677329540253,
|
|
"num_tokens": 2215239.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.2903533458709715,
|
|
"epoch": 0.9193357058125742,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 5.9998,
|
|
"mean_token_accuracy": 0.1248672790825367,
|
|
"num_tokens": 2229369.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.169124412536621,
|
|
"epoch": 0.9252669039145908,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 5.9842,
|
|
"mean_token_accuracy": 0.12811334952712058,
|
|
"num_tokens": 2242709.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.2337898254394535,
|
|
"epoch": 0.9311981020166074,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.12137661054730416,
|
|
"num_tokens": 2257067.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.1931390285491945,
|
|
"epoch": 0.9371293001186239,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.13152316808700562,
|
|
"num_tokens": 2270932.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.182427406311035,
|
|
"epoch": 0.9430604982206405,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.0279,
|
|
"mean_token_accuracy": 0.12549996599555016,
|
|
"num_tokens": 2285517.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.1117936134338375,
|
|
"epoch": 0.9489916963226572,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 5.8952,
|
|
"mean_token_accuracy": 0.1253364697098732,
|
|
"num_tokens": 2298657.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.14580602645874,
|
|
"epoch": 0.9549228944246738,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.132824869453907,
|
|
"num_tokens": 2312521.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.149325656890869,
|
|
"epoch": 0.9608540925266904,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 5.901,
|
|
"mean_token_accuracy": 0.12876850068569184,
|
|
"num_tokens": 2325849.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.160182809829712,
|
|
"epoch": 0.966785290628707,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 5.933,
|
|
"mean_token_accuracy": 0.12433248236775399,
|
|
"num_tokens": 2339717.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.148682689666748,
|
|
"epoch": 0.9727164887307236,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 5.9644,
|
|
"mean_token_accuracy": 0.12969405129551886,
|
|
"num_tokens": 2353536.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.15325140953064,
|
|
"epoch": 0.9786476868327402,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 5.9392,
|
|
"mean_token_accuracy": 0.12233664244413375,
|
|
"num_tokens": 2368626.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.148900079727173,
|
|
"epoch": 0.9845788849347569,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.0228,
|
|
"mean_token_accuracy": 0.1229950562119484,
|
|
"num_tokens": 2383768.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.253143930435181,
|
|
"epoch": 0.9905100830367735,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.0531,
|
|
"mean_token_accuracy": 0.12633627876639367,
|
|
"num_tokens": 2398089.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.143814277648926,
|
|
"epoch": 0.99644128113879,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 5.9179,
|
|
"mean_token_accuracy": 0.13075150698423385,
|
|
"num_tokens": 2411354.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.1597785472869875,
|
|
"epoch": 1.0023724792408066,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 5.9132,
|
|
"mean_token_accuracy": 0.1338951252400875,
|
|
"num_tokens": 2425071.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.226427221298218,
|
|
"epoch": 1.0083036773428233,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 5.8618,
|
|
"mean_token_accuracy": 0.13656074777245522,
|
|
"num_tokens": 2439143.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 5.997689485549927,
|
|
"epoch": 1.0142348754448398,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 5.7022,
|
|
"mean_token_accuracy": 0.1348326951265335,
|
|
"num_tokens": 2453466.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.064882230758667,
|
|
"epoch": 1.0201660735468565,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 5.815,
|
|
"mean_token_accuracy": 0.13819923251867294,
|
|
"num_tokens": 2466802.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.077611827850342,
|
|
"epoch": 1.026097271648873,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 5.7763,
|
|
"mean_token_accuracy": 0.13229055777192117,
|
|
"num_tokens": 2480080.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.0491447925567625,
|
|
"epoch": 1.0320284697508897,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 5.83,
|
|
"mean_token_accuracy": 0.1349202238023281,
|
|
"num_tokens": 2495196.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 5.99519829750061,
|
|
"epoch": 1.0379596678529064,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 5.734,
|
|
"mean_token_accuracy": 0.13301657065749167,
|
|
"num_tokens": 2510084.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.138531064987182,
|
|
"epoch": 1.0438908659549229,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 5.8045,
|
|
"mean_token_accuracy": 0.13315934240818023,
|
|
"num_tokens": 2522834.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.013190269470215,
|
|
"epoch": 1.0498220640569396,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 5.7781,
|
|
"mean_token_accuracy": 0.1408059172332287,
|
|
"num_tokens": 2536778.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.016625690460205,
|
|
"epoch": 1.055753262158956,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 5.8079,
|
|
"mean_token_accuracy": 0.13629683703184128,
|
|
"num_tokens": 2551220.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.152676439285278,
|
|
"epoch": 1.0616844602609727,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 5.8853,
|
|
"mean_token_accuracy": 0.12912147343158722,
|
|
"num_tokens": 2565686.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 5.9945775985717775,
|
|
"epoch": 1.0676156583629894,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 5.8012,
|
|
"mean_token_accuracy": 0.1301092065870762,
|
|
"num_tokens": 2579569.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.045119094848633,
|
|
"epoch": 1.073546856465006,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 5.7873,
|
|
"mean_token_accuracy": 0.13423676043748856,
|
|
"num_tokens": 2593226.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 5.971383666992187,
|
|
"epoch": 1.0794780545670226,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.13364143371582032,
|
|
"num_tokens": 2607711.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.028302955627441,
|
|
"epoch": 1.085409252669039,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 5.7788,
|
|
"mean_token_accuracy": 0.13531881719827651,
|
|
"num_tokens": 2623024.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.003001356124878,
|
|
"epoch": 1.0913404507710558,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 5.7666,
|
|
"mean_token_accuracy": 0.1361100286245346,
|
|
"num_tokens": 2637824.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 5.995635271072388,
|
|
"epoch": 1.0972716488730723,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 5.7772,
|
|
"mean_token_accuracy": 0.13633406162261963,
|
|
"num_tokens": 2652077.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 5.994116687774659,
|
|
"epoch": 1.103202846975089,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 5.7796,
|
|
"mean_token_accuracy": 0.1343744106590748,
|
|
"num_tokens": 2666791.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 5.9884919166564945,
|
|
"epoch": 1.1091340450771057,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 5.8653,
|
|
"mean_token_accuracy": 0.13153321593999862,
|
|
"num_tokens": 2681482.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 5.957932996749878,
|
|
"epoch": 1.1150652431791221,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 5.6686,
|
|
"mean_token_accuracy": 0.14378132447600364,
|
|
"num_tokens": 2695305.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.023182153701782,
|
|
"epoch": 1.1209964412811388,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 5.7336,
|
|
"mean_token_accuracy": 0.13786137402057647,
|
|
"num_tokens": 2712211.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 5.941826725006104,
|
|
"epoch": 1.1269276393831553,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 5.7816,
|
|
"mean_token_accuracy": 0.13546779826283456,
|
|
"num_tokens": 2726888.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 5.957491064071656,
|
|
"epoch": 1.132858837485172,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 5.7527,
|
|
"mean_token_accuracy": 0.13760311529040337,
|
|
"num_tokens": 2741734.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 5.94100170135498,
|
|
"epoch": 1.1387900355871885,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.14393994361162185,
|
|
"num_tokens": 2754680.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 5.94309253692627,
|
|
"epoch": 1.1447212336892052,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 5.8065,
|
|
"mean_token_accuracy": 0.129212536662817,
|
|
"num_tokens": 2769572.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.076050662994385,
|
|
"epoch": 1.150652431791222,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 5.7537,
|
|
"mean_token_accuracy": 0.13419156968593599,
|
|
"num_tokens": 2782948.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 5.846741008758545,
|
|
"epoch": 1.1565836298932384,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.6832,
|
|
"mean_token_accuracy": 0.1396567314863205,
|
|
"num_tokens": 2795399.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 5.9153892993927,
|
|
"epoch": 1.162514827995255,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.6837,
|
|
"mean_token_accuracy": 0.13754538744688033,
|
|
"num_tokens": 2809591.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 5.938817977905273,
|
|
"epoch": 1.1684460260972718,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 5.7227,
|
|
"mean_token_accuracy": 0.13820295482873918,
|
|
"num_tokens": 2824106.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 5.93713231086731,
|
|
"epoch": 1.1743772241992882,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 5.8459,
|
|
"mean_token_accuracy": 0.1346195101737976,
|
|
"num_tokens": 2838616.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 5.968208456039429,
|
|
"epoch": 1.180308422301305,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 5.7473,
|
|
"mean_token_accuracy": 0.13854797407984734,
|
|
"num_tokens": 2851707.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 5.898858261108399,
|
|
"epoch": 1.1862396204033214,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 5.7691,
|
|
"mean_token_accuracy": 0.1328132927417755,
|
|
"num_tokens": 2867644.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1862396204033214,
|
|
"eval_entropy": 5.91133521232137,
|
|
"eval_loss": 6.012734413146973,
|
|
"eval_mean_token_accuracy": 0.13139666496930313,
|
|
"eval_num_tokens": 2867644.0,
|
|
"eval_runtime": 2.4739,
|
|
"eval_samples_per_second": 1054.225,
|
|
"eval_steps_per_second": 131.778,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 5.986166191101074,
|
|
"epoch": 1.1921708185053381,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00049999967732573,
|
|
"loss": 5.7701,
|
|
"mean_token_accuracy": 0.1327154405415058,
|
|
"num_tokens": 2882663.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 5.940847110748291,
|
|
"epoch": 1.1981020166073546,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999983664630945,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.1329515464603901,
|
|
"num_tokens": 2898791.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 5.939939355850219,
|
|
"epoch": 1.2040332147093713,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999960472508219,
|
|
"loss": 5.7229,
|
|
"mean_token_accuracy": 0.14346506744623183,
|
|
"num_tokens": 2912997.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 5.890226602554321,
|
|
"epoch": 1.209964412811388,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999927196993059,
|
|
"loss": 5.7083,
|
|
"mean_token_accuracy": 0.1415116421878338,
|
|
"num_tokens": 2925064.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 5.959262037277222,
|
|
"epoch": 1.2158956109134045,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999883838234592,
|
|
"loss": 5.7633,
|
|
"mean_token_accuracy": 0.13465891852974893,
|
|
"num_tokens": 2941400.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 5.861705589294433,
|
|
"epoch": 1.2218268090154212,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999830396427133,
|
|
"loss": 5.7352,
|
|
"mean_token_accuracy": 0.13958436101675034,
|
|
"num_tokens": 2955541.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 5.922749280929565,
|
|
"epoch": 1.2277580071174377,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999766871810188,
|
|
"loss": 5.7479,
|
|
"mean_token_accuracy": 0.1416574686765671,
|
|
"num_tokens": 2968928.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 5.934699630737304,
|
|
"epoch": 1.2336892052194544,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999693264668446,
|
|
"loss": 5.6844,
|
|
"mean_token_accuracy": 0.14234655275940894,
|
|
"num_tokens": 2984072.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 5.891662549972534,
|
|
"epoch": 1.2396204033214708,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999609575331786,
|
|
"loss": 5.7787,
|
|
"mean_token_accuracy": 0.13772650435566902,
|
|
"num_tokens": 2999083.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 5.854400157928467,
|
|
"epoch": 1.2455516014234875,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999515804175268,
|
|
"loss": 5.726,
|
|
"mean_token_accuracy": 0.13745229691267014,
|
|
"num_tokens": 3012738.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 5.869228887557983,
|
|
"epoch": 1.2514827995255042,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999411951619134,
|
|
"loss": 5.6694,
|
|
"mean_token_accuracy": 0.14171827286481858,
|
|
"num_tokens": 3027929.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 5.864783573150635,
|
|
"epoch": 1.2574139976275207,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000499929801812881,
|
|
"loss": 5.8025,
|
|
"mean_token_accuracy": 0.13281177058815957,
|
|
"num_tokens": 3043576.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 5.822867107391358,
|
|
"epoch": 1.2633451957295374,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999174004214899,
|
|
"loss": 5.6696,
|
|
"mean_token_accuracy": 0.1398370385169983,
|
|
"num_tokens": 3059352.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 5.936639642715454,
|
|
"epoch": 1.269276393831554,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999039910433179,
|
|
"loss": 5.7859,
|
|
"mean_token_accuracy": 0.1422264978289604,
|
|
"num_tokens": 3073140.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 5.890475559234619,
|
|
"epoch": 1.2752075919335706,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998895737384605,
|
|
"loss": 5.7601,
|
|
"mean_token_accuracy": 0.14164954274892808,
|
|
"num_tokens": 3087710.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 5.919763278961182,
|
|
"epoch": 1.281138790035587,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998741485715297,
|
|
"loss": 5.713,
|
|
"mean_token_accuracy": 0.14089994207024575,
|
|
"num_tokens": 3102035.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 5.838283920288086,
|
|
"epoch": 1.2870699881376038,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998577156116551,
|
|
"loss": 5.7754,
|
|
"mean_token_accuracy": 0.13512811809778214,
|
|
"num_tokens": 3118534.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 5.856381511688232,
|
|
"epoch": 1.2930011862396205,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004998402749324823,
|
|
"loss": 5.7229,
|
|
"mean_token_accuracy": 0.13245076537132264,
|
|
"num_tokens": 3133419.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 5.997888469696045,
|
|
"epoch": 1.298932384341637,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004998218266121731,
|
|
"loss": 5.8243,
|
|
"mean_token_accuracy": 0.1440747559070587,
|
|
"num_tokens": 3145903.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 5.824967718124389,
|
|
"epoch": 1.3048635824436536,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998023707334055,
|
|
"loss": 5.7688,
|
|
"mean_token_accuracy": 0.1350715659558773,
|
|
"num_tokens": 3159938.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 5.8246561050415036,
|
|
"epoch": 1.3107947805456703,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997819073833724,
|
|
"loss": 5.6475,
|
|
"mean_token_accuracy": 0.1452900804579258,
|
|
"num_tokens": 3175089.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 5.855417728424072,
|
|
"epoch": 1.3167259786476868,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997604366537823,
|
|
"loss": 5.7858,
|
|
"mean_token_accuracy": 0.13705137223005295,
|
|
"num_tokens": 3191027.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 5.746766853332519,
|
|
"epoch": 1.3226571767497035,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499737958640858,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.14565233290195465,
|
|
"num_tokens": 3205489.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 5.841010189056396,
|
|
"epoch": 1.32858837485172,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997144734453367,
|
|
"loss": 5.604,
|
|
"mean_token_accuracy": 0.14238353818655014,
|
|
"num_tokens": 3218764.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 5.837244129180908,
|
|
"epoch": 1.3345195729537367,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996899811724693,
|
|
"loss": 5.7391,
|
|
"mean_token_accuracy": 0.13837844729423524,
|
|
"num_tokens": 3232953.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 5.772788858413696,
|
|
"epoch": 1.3404507710557532,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996644819320201,
|
|
"loss": 5.646,
|
|
"mean_token_accuracy": 0.14131752848625184,
|
|
"num_tokens": 3246977.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 5.676492929458618,
|
|
"epoch": 1.3463819691577699,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004996379758382662,
|
|
"loss": 5.5467,
|
|
"mean_token_accuracy": 0.14851859956979752,
|
|
"num_tokens": 3260668.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 5.83406023979187,
|
|
"epoch": 1.3523131672597866,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996104630099969,
|
|
"loss": 5.6622,
|
|
"mean_token_accuracy": 0.1436814546585083,
|
|
"num_tokens": 3274213.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 5.770905447006226,
|
|
"epoch": 1.358244365361803,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995819435705133,
|
|
"loss": 5.6455,
|
|
"mean_token_accuracy": 0.13666117042303086,
|
|
"num_tokens": 3288011.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 5.856086683273316,
|
|
"epoch": 1.3641755634638197,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995524176476277,
|
|
"loss": 5.7079,
|
|
"mean_token_accuracy": 0.1383526973426342,
|
|
"num_tokens": 3303263.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 5.795949029922485,
|
|
"epoch": 1.3701067615658362,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995218853736631,
|
|
"loss": 5.6848,
|
|
"mean_token_accuracy": 0.14614594876766204,
|
|
"num_tokens": 3316843.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 5.752928686141968,
|
|
"epoch": 1.376037959667853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994903468854527,
|
|
"loss": 5.5688,
|
|
"mean_token_accuracy": 0.14343352913856505,
|
|
"num_tokens": 3331527.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 5.751480293273926,
|
|
"epoch": 1.3819691577698694,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994578023243386,
|
|
"loss": 5.6655,
|
|
"mean_token_accuracy": 0.1439042344689369,
|
|
"num_tokens": 3347795.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 5.797406005859375,
|
|
"epoch": 1.387900355871886,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004994242518361723,
|
|
"loss": 5.7132,
|
|
"mean_token_accuracy": 0.1414179876446724,
|
|
"num_tokens": 3363532.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.777079343795776,
|
|
"epoch": 1.3938315539739028,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993896955713133,
|
|
"loss": 5.6743,
|
|
"mean_token_accuracy": 0.14409229755401612,
|
|
"num_tokens": 3377040.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.753640937805176,
|
|
"epoch": 1.3997627520759193,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004993541336846283,
|
|
"loss": 5.6287,
|
|
"mean_token_accuracy": 0.1529825061559677,
|
|
"num_tokens": 3390663.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.791894626617432,
|
|
"epoch": 1.405693950177936,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499317566335491,
|
|
"loss": 5.6625,
|
|
"mean_token_accuracy": 0.1494051843881607,
|
|
"num_tokens": 3405860.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.734993839263916,
|
|
"epoch": 1.4116251482799527,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992799936877813,
|
|
"loss": 5.6489,
|
|
"mean_token_accuracy": 0.14007329940795898,
|
|
"num_tokens": 3420248.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 5.817593097686768,
|
|
"epoch": 1.4175563463819691,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004992414159098841,
|
|
"loss": 5.6469,
|
|
"mean_token_accuracy": 0.14704996794462205,
|
|
"num_tokens": 3434126.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.779922151565552,
|
|
"epoch": 1.4234875444839858,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004992018331746891,
|
|
"loss": 5.7038,
|
|
"mean_token_accuracy": 0.14136891365051268,
|
|
"num_tokens": 3449462.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 5.80670485496521,
|
|
"epoch": 1.4294187425860023,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991612456595896,
|
|
"loss": 5.7589,
|
|
"mean_token_accuracy": 0.1408810332417488,
|
|
"num_tokens": 3465536.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.7316718101501465,
|
|
"epoch": 1.435349940688019,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991196535464824,
|
|
"loss": 5.6359,
|
|
"mean_token_accuracy": 0.14721256196498872,
|
|
"num_tokens": 3479129.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.719354104995728,
|
|
"epoch": 1.4412811387900355,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004990770570217659,
|
|
"loss": 5.5441,
|
|
"mean_token_accuracy": 0.15703994482755662,
|
|
"num_tokens": 3492191.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.779579067230225,
|
|
"epoch": 1.4472123368920522,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004990334562763401,
|
|
"loss": 5.7006,
|
|
"mean_token_accuracy": 0.1453902617096901,
|
|
"num_tokens": 3507459.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.741008663177491,
|
|
"epoch": 1.453143534994069,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004989888515056057,
|
|
"loss": 5.6628,
|
|
"mean_token_accuracy": 0.14069193080067635,
|
|
"num_tokens": 3521517.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.746186399459839,
|
|
"epoch": 1.4590747330960854,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004989432429094627,
|
|
"loss": 5.6036,
|
|
"mean_token_accuracy": 0.1472635604441166,
|
|
"num_tokens": 3535279.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.725585222244263,
|
|
"epoch": 1.465005931198102,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004988966306923099,
|
|
"loss": 5.5219,
|
|
"mean_token_accuracy": 0.1509362980723381,
|
|
"num_tokens": 3549960.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.702635431289673,
|
|
"epoch": 1.4709371293001186,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004988490150630441,
|
|
"loss": 5.6504,
|
|
"mean_token_accuracy": 0.1413729466497898,
|
|
"num_tokens": 3563814.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.836396932601929,
|
|
"epoch": 1.4768683274021353,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988003962350589,
|
|
"loss": 5.653,
|
|
"mean_token_accuracy": 0.15044173300266267,
|
|
"num_tokens": 3577529.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.712747097015381,
|
|
"epoch": 1.4827995255041517,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004987507744262436,
|
|
"loss": 5.5913,
|
|
"mean_token_accuracy": 0.1456787832081318,
|
|
"num_tokens": 3591812.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.635834550857544,
|
|
"epoch": 1.4887307236061684,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987001498589828,
|
|
"loss": 5.5787,
|
|
"mean_token_accuracy": 0.14813287556171417,
|
|
"num_tokens": 3604954.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.760722827911377,
|
|
"epoch": 1.4946619217081851,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986485227601548,
|
|
"loss": 5.6178,
|
|
"mean_token_accuracy": 0.14130520299077035,
|
|
"num_tokens": 3620486.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.717405652999878,
|
|
"epoch": 1.5005931198102016,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000498595893361131,
|
|
"loss": 5.555,
|
|
"mean_token_accuracy": 0.15612171441316605,
|
|
"num_tokens": 3633972.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.7364737033844,
|
|
"epoch": 1.5065243179122183,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004985422618977744,
|
|
"loss": 5.6037,
|
|
"mean_token_accuracy": 0.151451975107193,
|
|
"num_tokens": 3649306.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.7654664516448975,
|
|
"epoch": 1.512455516014235,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004984876286104392,
|
|
"loss": 5.6786,
|
|
"mean_token_accuracy": 0.14375052377581596,
|
|
"num_tokens": 3664809.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.70614218711853,
|
|
"epoch": 1.5183867141162515,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004984319937439693,
|
|
"loss": 5.6095,
|
|
"mean_token_accuracy": 0.15040914267301558,
|
|
"num_tokens": 3677669.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.743654489517212,
|
|
"epoch": 1.524317912218268,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983753575476969,
|
|
"loss": 5.6267,
|
|
"mean_token_accuracy": 0.14291197434067726,
|
|
"num_tokens": 3690588.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.696566772460938,
|
|
"epoch": 1.5302491103202847,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004983177202754421,
|
|
"loss": 5.6522,
|
|
"mean_token_accuracy": 0.14571921825408934,
|
|
"num_tokens": 3703749.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.805274868011475,
|
|
"epoch": 1.5361803084223014,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004982590821855115,
|
|
"loss": 5.68,
|
|
"mean_token_accuracy": 0.13761823400855064,
|
|
"num_tokens": 3720181.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.711960506439209,
|
|
"epoch": 1.5421115065243178,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004981994435406965,
|
|
"loss": 5.6042,
|
|
"mean_token_accuracy": 0.14782098084688186,
|
|
"num_tokens": 3734918.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.666572904586792,
|
|
"epoch": 1.5480427046263345,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004981388046082731,
|
|
"loss": 5.5494,
|
|
"mean_token_accuracy": 0.15440027564764022,
|
|
"num_tokens": 3748660.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.685553789138794,
|
|
"epoch": 1.5539739027283512,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004980771656599998,
|
|
"loss": 5.6201,
|
|
"mean_token_accuracy": 0.14841450601816178,
|
|
"num_tokens": 3763481.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.763680267333984,
|
|
"epoch": 1.5599051008303677,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004980145269721168,
|
|
"loss": 5.6376,
|
|
"mean_token_accuracy": 0.14404506087303162,
|
|
"num_tokens": 3777077.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.685877466201783,
|
|
"epoch": 1.5658362989323842,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000497950888825345,
|
|
"loss": 5.6386,
|
|
"mean_token_accuracy": 0.14266225099563598,
|
|
"num_tokens": 3793649.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.687498426437378,
|
|
"epoch": 1.571767497034401,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000497886251504884,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.15134050250053405,
|
|
"num_tokens": 3807712.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.639243650436401,
|
|
"epoch": 1.5776986951364176,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004978206153004117,
|
|
"loss": 5.5347,
|
|
"mean_token_accuracy": 0.1552049770951271,
|
|
"num_tokens": 3821637.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.8086631298065186,
|
|
"epoch": 1.583629893238434,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977539805060824,
|
|
"loss": 5.6912,
|
|
"mean_token_accuracy": 0.1436895474791527,
|
|
"num_tokens": 3836272.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.614665985107422,
|
|
"epoch": 1.5895610913404508,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004976863474205257,
|
|
"loss": 5.5055,
|
|
"mean_token_accuracy": 0.15429906994104386,
|
|
"num_tokens": 3850018.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.774603271484375,
|
|
"epoch": 1.5954922894424675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976177163468451,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.15020062923431396,
|
|
"num_tokens": 3865061.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.631220006942749,
|
|
"epoch": 1.601423487544484,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004975480875926168,
|
|
"loss": 5.6835,
|
|
"mean_token_accuracy": 0.15097753554582596,
|
|
"num_tokens": 3879697.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.735270404815674,
|
|
"epoch": 1.6073546856465006,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000497477461469888,
|
|
"loss": 5.4825,
|
|
"mean_token_accuracy": 0.15924198776483536,
|
|
"num_tokens": 3891889.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.665686845779419,
|
|
"epoch": 1.6132858837485173,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974058382951757,
|
|
"loss": 5.5945,
|
|
"mean_token_accuracy": 0.1450992450118065,
|
|
"num_tokens": 3907356.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.731923580169678,
|
|
"epoch": 1.6192170818505338,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004973332183894653,
|
|
"loss": 5.5786,
|
|
"mean_token_accuracy": 0.14734622836112976,
|
|
"num_tokens": 3921839.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.587240982055664,
|
|
"epoch": 1.6251482799525503,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004972596020782096,
|
|
"loss": 5.483,
|
|
"mean_token_accuracy": 0.15152351856231688,
|
|
"num_tokens": 3937266.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.740517473220825,
|
|
"epoch": 1.631079478054567,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004971849896913259,
|
|
"loss": 5.6174,
|
|
"mean_token_accuracy": 0.14421468004584312,
|
|
"num_tokens": 3952721.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.699207830429077,
|
|
"epoch": 1.6370106761565837,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004971093815631965,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.14614131152629853,
|
|
"num_tokens": 3966557.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.577703142166138,
|
|
"epoch": 1.6429418742586002,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004970327780326655,
|
|
"loss": 5.4592,
|
|
"mean_token_accuracy": 0.15435569435358049,
|
|
"num_tokens": 3978739.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.523289442062378,
|
|
"epoch": 1.6488730723606169,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004969551794430383,
|
|
"loss": 5.4758,
|
|
"mean_token_accuracy": 0.15182152688503264,
|
|
"num_tokens": 3993868.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.735844373703003,
|
|
"epoch": 1.6548042704626336,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004968765861420799,
|
|
"loss": 5.6032,
|
|
"mean_token_accuracy": 0.15374974012374878,
|
|
"num_tokens": 4006661.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.63426947593689,
|
|
"epoch": 1.66073546856465,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004967969984820128,
|
|
"loss": 5.5828,
|
|
"mean_token_accuracy": 0.1495743677020073,
|
|
"num_tokens": 4021670.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.673282051086426,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004967164168195159,
|
|
"loss": 5.5646,
|
|
"mean_token_accuracy": 0.1479659453034401,
|
|
"num_tokens": 4036995.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.667166376113892,
|
|
"epoch": 1.6725978647686834,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000496634841515723,
|
|
"loss": 5.6035,
|
|
"mean_token_accuracy": 0.15370306074619294,
|
|
"num_tokens": 4049464.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.649252462387085,
|
|
"epoch": 1.6785290628707,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004965522729362209,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.15451542884111405,
|
|
"num_tokens": 4063630.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.63691987991333,
|
|
"epoch": 1.6844602609727164,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004964687114510478,
|
|
"loss": 5.5176,
|
|
"mean_token_accuracy": 0.14818612933158876,
|
|
"num_tokens": 4079890.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.556159973144531,
|
|
"epoch": 1.690391459074733,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004963841574346917,
|
|
"loss": 5.5784,
|
|
"mean_token_accuracy": 0.15199574530124665,
|
|
"num_tokens": 4095693.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.76162371635437,
|
|
"epoch": 1.6963226571767498,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004962986112660887,
|
|
"loss": 5.6264,
|
|
"mean_token_accuracy": 0.14758573174476625,
|
|
"num_tokens": 4109655.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.5980674743652346,
|
|
"epoch": 1.7022538552787663,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004962120733286215,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.15562087446451187,
|
|
"num_tokens": 4123870.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.724174547195434,
|
|
"epoch": 1.708185053380783,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004961245440101171,
|
|
"loss": 5.6127,
|
|
"mean_token_accuracy": 0.15370832234621049,
|
|
"num_tokens": 4138009.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.661290311813355,
|
|
"epoch": 1.7141162514827997,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000496036023702846,
|
|
"loss": 5.6985,
|
|
"mean_token_accuracy": 0.14476853236556053,
|
|
"num_tokens": 4151929.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.681108665466309,
|
|
"epoch": 1.7200474495848161,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004959465128035195,
|
|
"loss": 5.5322,
|
|
"mean_token_accuracy": 0.1587254822254181,
|
|
"num_tokens": 4164626.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.668149662017822,
|
|
"epoch": 1.7259786476868326,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004958560117132884,
|
|
"loss": 5.4829,
|
|
"mean_token_accuracy": 0.1552940219640732,
|
|
"num_tokens": 4179418.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.586788272857666,
|
|
"epoch": 1.7319098457888493,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004957645208377411,
|
|
"loss": 5.5927,
|
|
"mean_token_accuracy": 0.14715088456869124,
|
|
"num_tokens": 4194385.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.677043342590332,
|
|
"epoch": 1.737841043890866,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000495672040586902,
|
|
"loss": 5.5726,
|
|
"mean_token_accuracy": 0.1476666510105133,
|
|
"num_tokens": 4209246.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.573235273361206,
|
|
"epoch": 1.7437722419928825,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004955785713752292,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.1514071211218834,
|
|
"num_tokens": 4225142.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.6124965190887455,
|
|
"epoch": 1.7497034400948992,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004954841136216132,
|
|
"loss": 5.5121,
|
|
"mean_token_accuracy": 0.1569083496928215,
|
|
"num_tokens": 4239495.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.6624431133270265,
|
|
"epoch": 1.755634638196916,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953886677493746,
|
|
"loss": 5.4459,
|
|
"mean_token_accuracy": 0.16436467915773392,
|
|
"num_tokens": 4251780.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.616982173919678,
|
|
"epoch": 1.7615658362989324,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004952922341862622,
|
|
"loss": 5.6927,
|
|
"mean_token_accuracy": 0.14587103947997093,
|
|
"num_tokens": 4265490.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.64134464263916,
|
|
"epoch": 1.7674970344009489,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004951948133644514,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.15793099850416184,
|
|
"num_tokens": 4280478.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.635649013519287,
|
|
"epoch": 1.7734282325029656,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004950964057205421,
|
|
"loss": 5.5551,
|
|
"mean_token_accuracy": 0.1520090416073799,
|
|
"num_tokens": 4296649.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.565133810043335,
|
|
"epoch": 1.7793594306049823,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004949970116955566,
|
|
"loss": 5.525,
|
|
"mean_token_accuracy": 0.15610957741737366,
|
|
"num_tokens": 4310470.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7793594306049823,
|
|
"eval_entropy": 5.4890358755193605,
|
|
"eval_loss": 5.720747947692871,
|
|
"eval_mean_token_accuracy": 0.15074070664179837,
|
|
"eval_num_tokens": 4310470.0,
|
|
"eval_runtime": 2.4551,
|
|
"eval_samples_per_second": 1062.291,
|
|
"eval_steps_per_second": 132.786,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.62813172340393,
|
|
"epoch": 1.7852906287069987,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004948966317349379,
|
|
"loss": 5.5502,
|
|
"mean_token_accuracy": 0.15073443800210953,
|
|
"num_tokens": 4324996.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.527570104598999,
|
|
"epoch": 1.7912218268090154,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004947952662885473,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.15607783049345017,
|
|
"num_tokens": 4337194.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.547454500198365,
|
|
"epoch": 1.7971530249110321,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000494692915810663,
|
|
"loss": 5.4743,
|
|
"mean_token_accuracy": 0.1528942972421646,
|
|
"num_tokens": 4352516.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.56812629699707,
|
|
"epoch": 1.8030842230130486,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004945895807599775,
|
|
"loss": 5.5122,
|
|
"mean_token_accuracy": 0.16002019345760346,
|
|
"num_tokens": 4367654.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.605467319488525,
|
|
"epoch": 1.809015421115065,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004944852615995956,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.15724539756774902,
|
|
"num_tokens": 4381275.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.572732782363891,
|
|
"epoch": 1.814946619217082,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004943799587970328,
|
|
"loss": 5.494,
|
|
"mean_token_accuracy": 0.1523606076836586,
|
|
"num_tokens": 4394392.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.530032682418823,
|
|
"epoch": 1.8208778173190985,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004942736728242127,
|
|
"loss": 5.4026,
|
|
"mean_token_accuracy": 0.15689572244882583,
|
|
"num_tokens": 4407512.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.584712648391724,
|
|
"epoch": 1.826809015421115,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004941664041574651,
|
|
"loss": 5.507,
|
|
"mean_token_accuracy": 0.15881744623184205,
|
|
"num_tokens": 4422223.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.6050506114959715,
|
|
"epoch": 1.8327402135231317,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004940581532775237,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.1518533781170845,
|
|
"num_tokens": 4436994.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.4888392925262455,
|
|
"epoch": 1.8386714116251484,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004939489206695241,
|
|
"loss": 5.4267,
|
|
"mean_token_accuracy": 0.15675119161605836,
|
|
"num_tokens": 4450429.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.62486629486084,
|
|
"epoch": 1.8446026097271648,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004938387068230021,
|
|
"loss": 5.5789,
|
|
"mean_token_accuracy": 0.15119255930185319,
|
|
"num_tokens": 4465991.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.55673360824585,
|
|
"epoch": 1.8505338078291815,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004937275122318902,
|
|
"loss": 5.4977,
|
|
"mean_token_accuracy": 0.15660169571638108,
|
|
"num_tokens": 4480873.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.634980964660644,
|
|
"epoch": 1.8564650059311982,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004936153373945165,
|
|
"loss": 5.5428,
|
|
"mean_token_accuracy": 0.159622260928154,
|
|
"num_tokens": 4493443.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.580209970474243,
|
|
"epoch": 1.8623962040332147,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004935021828136025,
|
|
"loss": 5.5042,
|
|
"mean_token_accuracy": 0.15394536405801773,
|
|
"num_tokens": 4506457.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.547038459777832,
|
|
"epoch": 1.8683274021352312,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004933880489962602,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.15576548576354982,
|
|
"num_tokens": 4520529.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.558015251159668,
|
|
"epoch": 1.8742586002372479,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004932729364539901,
|
|
"loss": 5.3993,
|
|
"mean_token_accuracy": 0.15773553699254989,
|
|
"num_tokens": 4535100.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.509803533554077,
|
|
"epoch": 1.8801897983392646,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004931568457026789,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.1584371492266655,
|
|
"num_tokens": 4549152.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.5465106010437015,
|
|
"epoch": 1.886120996441281,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004930397772625976,
|
|
"loss": 5.3981,
|
|
"mean_token_accuracy": 0.16331460624933242,
|
|
"num_tokens": 4563302.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.470412969589233,
|
|
"epoch": 1.8920521945432978,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004929217316583985,
|
|
"loss": 5.3418,
|
|
"mean_token_accuracy": 0.1644173815846443,
|
|
"num_tokens": 4576770.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.56337342262268,
|
|
"epoch": 1.8979833926453145,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000492802709419113,
|
|
"loss": 5.4922,
|
|
"mean_token_accuracy": 0.1560358837246895,
|
|
"num_tokens": 4590264.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.541475391387939,
|
|
"epoch": 1.903914590747331,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004926827110781496,
|
|
"loss": 5.4658,
|
|
"mean_token_accuracy": 0.1521800145506859,
|
|
"num_tokens": 4606217.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.499109220504761,
|
|
"epoch": 1.9098457888493474,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004925617371732914,
|
|
"loss": 5.3749,
|
|
"mean_token_accuracy": 0.16113989651203156,
|
|
"num_tokens": 4620460.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.545439529418945,
|
|
"epoch": 1.9157769869513643,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004924397882466933,
|
|
"loss": 5.4441,
|
|
"mean_token_accuracy": 0.15975887924432755,
|
|
"num_tokens": 4635915.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.561807680130005,
|
|
"epoch": 1.9217081850533808,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004923168648448799,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.15793247371912003,
|
|
"num_tokens": 4650789.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.594422292709351,
|
|
"epoch": 1.9276393831553973,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004921929675187433,
|
|
"loss": 5.518,
|
|
"mean_token_accuracy": 0.15074024945497513,
|
|
"num_tokens": 4666306.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.519123840332031,
|
|
"epoch": 1.933570581257414,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004920680968235397,
|
|
"loss": 5.5125,
|
|
"mean_token_accuracy": 0.1595379963517189,
|
|
"num_tokens": 4679587.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.621131181716919,
|
|
"epoch": 1.9395017793594307,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004919422533188882,
|
|
"loss": 5.441,
|
|
"mean_token_accuracy": 0.15777284204959868,
|
|
"num_tokens": 4694237.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.571220350265503,
|
|
"epoch": 1.9454329774614472,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004918154375687673,
|
|
"loss": 5.5383,
|
|
"mean_token_accuracy": 0.1469981402158737,
|
|
"num_tokens": 4708423.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.614190530776978,
|
|
"epoch": 1.9513641755634639,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004916876501415129,
|
|
"loss": 5.495,
|
|
"mean_token_accuracy": 0.15826049596071243,
|
|
"num_tokens": 4723039.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.505533838272095,
|
|
"epoch": 1.9572953736654806,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004915588916098151,
|
|
"loss": 5.4317,
|
|
"mean_token_accuracy": 0.16039896309375762,
|
|
"num_tokens": 4735882.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.477867507934571,
|
|
"epoch": 1.963226571767497,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004914291625507164,
|
|
"loss": 5.4436,
|
|
"mean_token_accuracy": 0.16591833531856537,
|
|
"num_tokens": 4748810.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.561583375930786,
|
|
"epoch": 1.9691577698695135,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000491298463545609,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.1546772077679634,
|
|
"num_tokens": 4763563.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.511043453216553,
|
|
"epoch": 1.9750889679715302,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004911667951802318,
|
|
"loss": 5.4528,
|
|
"mean_token_accuracy": 0.15551470816135407,
|
|
"num_tokens": 4775483.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.501114273071289,
|
|
"epoch": 1.981020166073547,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000491034158044668,
|
|
"loss": 5.4583,
|
|
"mean_token_accuracy": 0.1569467306137085,
|
|
"num_tokens": 4792507.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.568025827407837,
|
|
"epoch": 1.9869513641755634,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004909005527333421,
|
|
"loss": 5.5004,
|
|
"mean_token_accuracy": 0.155551216006279,
|
|
"num_tokens": 4807722.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.516203594207764,
|
|
"epoch": 1.99288256227758,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000490765979845018,
|
|
"loss": 5.32,
|
|
"mean_token_accuracy": 0.16825991868972778,
|
|
"num_tokens": 4821129.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.505976819992066,
|
|
"epoch": 1.9988137603795968,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000490630439982796,
|
|
"loss": 5.3815,
|
|
"mean_token_accuracy": 0.15964729189872742,
|
|
"num_tokens": 4835483.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.534174060821533,
|
|
"epoch": 2.0047449584816133,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004904939337541093,
|
|
"loss": 5.165,
|
|
"mean_token_accuracy": 0.17265937626361846,
|
|
"num_tokens": 4848607.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.4161741733551025,
|
|
"epoch": 2.0106761565836297,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004903564617707227,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.17384703904390336,
|
|
"num_tokens": 4863324.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.535564327239991,
|
|
"epoch": 2.0166073546856467,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004902180246487285,
|
|
"loss": 5.1794,
|
|
"mean_token_accuracy": 0.17566659599542617,
|
|
"num_tokens": 4876603.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.487852954864502,
|
|
"epoch": 2.022538552787663,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000490078623008545,
|
|
"loss": 5.2708,
|
|
"mean_token_accuracy": 0.16383918672800063,
|
|
"num_tokens": 4893733.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.522866249084473,
|
|
"epoch": 2.0284697508896796,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004899382574749124,
|
|
"loss": 5.2032,
|
|
"mean_token_accuracy": 0.1683608591556549,
|
|
"num_tokens": 4908476.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.46659140586853,
|
|
"epoch": 2.0344009489916965,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000489796928676891,
|
|
"loss": 5.1535,
|
|
"mean_token_accuracy": 0.17724859714508057,
|
|
"num_tokens": 4920471.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.461789083480835,
|
|
"epoch": 2.040332147093713,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004896546372478581,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.17168280780315398,
|
|
"num_tokens": 4934984.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.375751543045044,
|
|
"epoch": 2.0462633451957295,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004895113838255049,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.16719358563423156,
|
|
"num_tokens": 4949709.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.503344297409058,
|
|
"epoch": 2.052194543297746,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004893671690518343,
|
|
"loss": 5.1678,
|
|
"mean_token_accuracy": 0.17527552992105483,
|
|
"num_tokens": 4964439.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.424249362945557,
|
|
"epoch": 2.058125741399763,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004892219935731571,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.1706253096461296,
|
|
"num_tokens": 4979958.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.445143413543701,
|
|
"epoch": 2.0640569395017794,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004890758580400898,
|
|
"loss": 5.2107,
|
|
"mean_token_accuracy": 0.17454935014247894,
|
|
"num_tokens": 4993034.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.3907262802124025,
|
|
"epoch": 2.069988137603796,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004889287631075513,
|
|
"loss": 5.1469,
|
|
"mean_token_accuracy": 0.17527274340391158,
|
|
"num_tokens": 5008580.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.415247631072998,
|
|
"epoch": 2.0759193357058128,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004887807094347606,
|
|
"loss": 5.1757,
|
|
"mean_token_accuracy": 0.1709846779704094,
|
|
"num_tokens": 5022322.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.5165040493011475,
|
|
"epoch": 2.0818505338078293,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004886316976852326,
|
|
"loss": 5.1905,
|
|
"mean_token_accuracy": 0.16880395114421845,
|
|
"num_tokens": 5036511.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.393788051605225,
|
|
"epoch": 2.0877817319098457,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004884817285267766,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.1742696449160576,
|
|
"num_tokens": 5050129.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.337964820861816,
|
|
"epoch": 2.093712930011862,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004883308026314923,
|
|
"loss": 5.0561,
|
|
"mean_token_accuracy": 0.18535924702882767,
|
|
"num_tokens": 5062299.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.378459167480469,
|
|
"epoch": 2.099644128113879,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048817892067576726,
|
|
"loss": 5.1866,
|
|
"mean_token_accuracy": 0.17031457126140595,
|
|
"num_tokens": 5077665.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.357861471176148,
|
|
"epoch": 2.1055753262158956,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004880260833402734,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.1715405285358429,
|
|
"num_tokens": 5092005.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.379746150970459,
|
|
"epoch": 2.111506524317912,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004878722913099645,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.16930181831121444,
|
|
"num_tokens": 5109155.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.383475065231323,
|
|
"epoch": 2.117437722419929,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000487717545274073,
|
|
"loss": 5.149,
|
|
"mean_token_accuracy": 0.16863701194524766,
|
|
"num_tokens": 5125791.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.384926605224609,
|
|
"epoch": 2.1233689205219455,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004875618459261065,
|
|
"loss": 5.1762,
|
|
"mean_token_accuracy": 0.16971782743930816,
|
|
"num_tokens": 5139908.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.391865921020508,
|
|
"epoch": 2.129300118623962,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048740519396384515,
|
|
"loss": 5.1527,
|
|
"mean_token_accuracy": 0.17617459148168563,
|
|
"num_tokens": 5154922.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.33276104927063,
|
|
"epoch": 2.135231316725979,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004872475900893382,
|
|
"loss": 5.1899,
|
|
"mean_token_accuracy": 0.17570777982473373,
|
|
"num_tokens": 5169919.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.48026123046875,
|
|
"epoch": 2.1411625148279954,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004870890350089011,
|
|
"loss": 5.1829,
|
|
"mean_token_accuracy": 0.16823343187570572,
|
|
"num_tokens": 5186687.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.437422800064087,
|
|
"epoch": 2.147093712930012,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048692952943311214,
|
|
"loss": 5.2063,
|
|
"mean_token_accuracy": 0.1717593029141426,
|
|
"num_tokens": 5199489.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.496373796463013,
|
|
"epoch": 2.1530249110320283,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004867690740768094,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.17145588397979736,
|
|
"num_tokens": 5214549.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.366263675689697,
|
|
"epoch": 2.1589561091340452,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048660766965908736,
|
|
"loss": 5.2164,
|
|
"mean_token_accuracy": 0.16977577656507492,
|
|
"num_tokens": 5228571.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.388319301605224,
|
|
"epoch": 2.1648873072360617,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000486445316903294,
|
|
"loss": 5.1452,
|
|
"mean_token_accuracy": 0.17680991888046266,
|
|
"num_tokens": 5242797.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.426320600509643,
|
|
"epoch": 2.170818505338078,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000486282016537027,
|
|
"loss": 5.2368,
|
|
"mean_token_accuracy": 0.17110936492681503,
|
|
"num_tokens": 5257692.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.450334596633911,
|
|
"epoch": 2.176749703440095,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004861177692921313,
|
|
"loss": 5.2329,
|
|
"mean_token_accuracy": 0.17457327097654343,
|
|
"num_tokens": 5273117.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.369627332687378,
|
|
"epoch": 2.1826809015421116,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000485952575904695,
|
|
"loss": 5.1764,
|
|
"mean_token_accuracy": 0.165084633231163,
|
|
"num_tokens": 5287232.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.3623122215271,
|
|
"epoch": 2.188612099644128,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004857864371150467,
|
|
"loss": 5.164,
|
|
"mean_token_accuracy": 0.17174010574817658,
|
|
"num_tokens": 5300049.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.422196960449218,
|
|
"epoch": 2.1945432977461445,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004856193536677515,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.17225924283266067,
|
|
"num_tokens": 5313811.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.38545184135437,
|
|
"epoch": 2.2004744958481615,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048545132631160856,
|
|
"loss": 5.2414,
|
|
"mean_token_accuracy": 0.16925344467163086,
|
|
"num_tokens": 5326402.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.331728982925415,
|
|
"epoch": 2.206405693950178,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000485282355799647,
|
|
"loss": 5.1074,
|
|
"mean_token_accuracy": 0.1777871370315552,
|
|
"num_tokens": 5340440.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.373279333114624,
|
|
"epoch": 2.2123368920521944,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004851124428891226,
|
|
"loss": 5.2044,
|
|
"mean_token_accuracy": 0.17208765894174577,
|
|
"num_tokens": 5355271.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.3738720417022705,
|
|
"epoch": 2.2182680901542113,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048494158834151485,
|
|
"loss": 5.1657,
|
|
"mean_token_accuracy": 0.1768079251050949,
|
|
"num_tokens": 5371939.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.338125801086425,
|
|
"epoch": 2.224199288256228,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048476979292252323,
|
|
"loss": 5.0599,
|
|
"mean_token_accuracy": 0.18021448701620102,
|
|
"num_tokens": 5386242.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.32992787361145,
|
|
"epoch": 2.2301304863582443,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845970574020637,
|
|
"loss": 5.1392,
|
|
"mean_token_accuracy": 0.17196872234344482,
|
|
"num_tokens": 5399867.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.3204365253448485,
|
|
"epoch": 2.2360616844602608,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048442338255426555,
|
|
"loss": 5.0614,
|
|
"mean_token_accuracy": 0.1749480903148651,
|
|
"num_tokens": 5414236.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.360966062545776,
|
|
"epoch": 2.2419928825622777,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004842487691574675,
|
|
"loss": 5.1766,
|
|
"mean_token_accuracy": 0.17208879739046096,
|
|
"num_tokens": 5428510.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.411381673812866,
|
|
"epoch": 2.247924080664294,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004840732179942146,
|
|
"loss": 5.1757,
|
|
"mean_token_accuracy": 0.17247659713029861,
|
|
"num_tokens": 5443137.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.371445941925049,
|
|
"epoch": 2.2538552787663106,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048389672985125467,
|
|
"loss": 5.1897,
|
|
"mean_token_accuracy": 0.1735587641596794,
|
|
"num_tokens": 5456574.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.3538429737091064,
|
|
"epoch": 2.2597864768683276,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004837193055195344,
|
|
"loss": 5.1926,
|
|
"mean_token_accuracy": 0.17189497202634813,
|
|
"num_tokens": 5472789.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.422811889648438,
|
|
"epoch": 2.265717674970344,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048354094579419656,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.1655895009636879,
|
|
"num_tokens": 5488202.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.344785451889038,
|
|
"epoch": 2.2716488730723605,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048336165147457537,
|
|
"loss": 5.2373,
|
|
"mean_token_accuracy": 0.16763855069875716,
|
|
"num_tokens": 5502661.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.352733612060547,
|
|
"epoch": 2.277580071174377,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048318142336419413,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.17304945439100267,
|
|
"num_tokens": 5516371.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.342757987976074,
|
|
"epoch": 2.283511269276394,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004830002622707604,
|
|
"loss": 5.1239,
|
|
"mean_token_accuracy": 0.1793706387281418,
|
|
"num_tokens": 5529339.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.337879085540772,
|
|
"epoch": 2.2894424673784104,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048281816900616354,
|
|
"loss": 5.1158,
|
|
"mean_token_accuracy": 0.17504990845918655,
|
|
"num_tokens": 5543173.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.280277824401855,
|
|
"epoch": 2.295373665480427,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048263514438647015,
|
|
"loss": 5.1089,
|
|
"mean_token_accuracy": 0.17652505189180373,
|
|
"num_tokens": 5557080.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.356028890609741,
|
|
"epoch": 2.301304863582444,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004824511892319209,
|
|
"loss": 5.1702,
|
|
"mean_token_accuracy": 0.17574198096990584,
|
|
"num_tokens": 5571422.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.253097772598267,
|
|
"epoch": 2.3072360616844603,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004822663043669267,
|
|
"loss": 4.9664,
|
|
"mean_token_accuracy": 0.18942097276449205,
|
|
"num_tokens": 5583057.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.27740592956543,
|
|
"epoch": 2.3131672597864767,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048208049062006505,
|
|
"loss": 5.2096,
|
|
"mean_token_accuracy": 0.16950699239969252,
|
|
"num_tokens": 5596201.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.393002700805664,
|
|
"epoch": 2.3190984578884937,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004818937488240764,
|
|
"loss": 5.1601,
|
|
"mean_token_accuracy": 0.17385102808475494,
|
|
"num_tokens": 5612461.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.323347473144532,
|
|
"epoch": 2.32502965599051,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048170607981586023,
|
|
"loss": 5.2285,
|
|
"mean_token_accuracy": 0.16853837668895721,
|
|
"num_tokens": 5628148.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.288127470016479,
|
|
"epoch": 2.3309608540925266,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004815174844364714,
|
|
"loss": 5.0684,
|
|
"mean_token_accuracy": 0.1746617555618286,
|
|
"num_tokens": 5641695.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.371695184707642,
|
|
"epoch": 2.3368920521945435,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004813279635311165,
|
|
"loss": 5.1085,
|
|
"mean_token_accuracy": 0.18225748091936111,
|
|
"num_tokens": 5654902.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.291834688186645,
|
|
"epoch": 2.34282325029656,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048113751794914974,
|
|
"loss": 5.1022,
|
|
"mean_token_accuracy": 0.18446253687143327,
|
|
"num_tokens": 5666899.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.420601224899292,
|
|
"epoch": 2.3487544483985765,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004809461485440694,
|
|
"loss": 5.2152,
|
|
"mean_token_accuracy": 0.17578936219215394,
|
|
"num_tokens": 5682525.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.2827568531036375,
|
|
"epoch": 2.354685646500593,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004807538561735143,
|
|
"loss": 5.14,
|
|
"mean_token_accuracy": 0.16829978823661804,
|
|
"num_tokens": 5697097.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.2644284725189205,
|
|
"epoch": 2.36061684460261,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048056064169925904,
|
|
"loss": 5.0869,
|
|
"mean_token_accuracy": 0.17536696344614028,
|
|
"num_tokens": 5712340.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.353961038589477,
|
|
"epoch": 2.3665480427046264,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048036650598721125,
|
|
"loss": 5.197,
|
|
"mean_token_accuracy": 0.16865415275096893,
|
|
"num_tokens": 5726509.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.3073567867279055,
|
|
"epoch": 2.372479240806643,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048017144990740673,
|
|
"loss": 5.121,
|
|
"mean_token_accuracy": 0.1786181464791298,
|
|
"num_tokens": 5740977.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.372479240806643,
|
|
"eval_entropy": 5.170273498523454,
|
|
"eval_loss": 5.574867248535156,
|
|
"eval_mean_token_accuracy": 0.16265562777778853,
|
|
"eval_num_tokens": 5740977.0,
|
|
"eval_runtime": 2.4303,
|
|
"eval_samples_per_second": 1073.13,
|
|
"eval_steps_per_second": 134.141,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.288026952743531,
|
|
"epoch": 2.3784104389086593,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047997547433400615,
|
|
"loss": 5.0974,
|
|
"mean_token_accuracy": 0.1805935502052307,
|
|
"num_tokens": 5756000.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.338898134231568,
|
|
"epoch": 2.3843416370106763,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004797785801452911,
|
|
"loss": 5.1033,
|
|
"mean_token_accuracy": 0.18041265308856963,
|
|
"num_tokens": 5771438.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.329379606246948,
|
|
"epoch": 2.3902728351126927,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047958076822365994,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.17072732895612716,
|
|
"num_tokens": 5784686.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.344692611694336,
|
|
"epoch": 2.396204033214709,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004793820394556238,
|
|
"loss": 5.201,
|
|
"mean_token_accuracy": 0.17325544208288193,
|
|
"num_tokens": 5799363.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.308460712432861,
|
|
"epoch": 2.402135231316726,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004791823947318029,
|
|
"loss": 5.0887,
|
|
"mean_token_accuracy": 0.176689213514328,
|
|
"num_tokens": 5812109.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.2809264183044435,
|
|
"epoch": 2.4080664294187426,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004789818349469222,
|
|
"loss": 5.0932,
|
|
"mean_token_accuracy": 0.18039163053035737,
|
|
"num_tokens": 5825780.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.265401554107666,
|
|
"epoch": 2.413997627520759,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047878036099980795,
|
|
"loss": 5.1356,
|
|
"mean_token_accuracy": 0.16557523757219314,
|
|
"num_tokens": 5841184.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.28918867111206,
|
|
"epoch": 2.419928825622776,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047857797379338284,
|
|
"loss": 5.1151,
|
|
"mean_token_accuracy": 0.18016838133335114,
|
|
"num_tokens": 5854422.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.434517860412598,
|
|
"epoch": 2.4258600237247925,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047837467423466284,
|
|
"loss": 5.2756,
|
|
"mean_token_accuracy": 0.16469118744134903,
|
|
"num_tokens": 5870878.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.323330879211426,
|
|
"epoch": 2.431791221826809,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047817046323475246,
|
|
"loss": 5.2065,
|
|
"mean_token_accuracy": 0.16481388956308365,
|
|
"num_tokens": 5886137.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.346496486663819,
|
|
"epoch": 2.4377224199288254,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004779653417088408,
|
|
"loss": 5.1252,
|
|
"mean_token_accuracy": 0.17990775406360626,
|
|
"num_tokens": 5900300.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.322238492965698,
|
|
"epoch": 2.4436536180308424,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047775931057619786,
|
|
"loss": 5.2122,
|
|
"mean_token_accuracy": 0.1684417709708214,
|
|
"num_tokens": 5917933.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.431944942474365,
|
|
"epoch": 2.449584816132859,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047755237076016997,
|
|
"loss": 5.1549,
|
|
"mean_token_accuracy": 0.1798498421907425,
|
|
"num_tokens": 5932189.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.2723767280578615,
|
|
"epoch": 2.4555160142348753,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004773445231881757,
|
|
"loss": 5.1586,
|
|
"mean_token_accuracy": 0.17868874669075013,
|
|
"num_tokens": 5945999.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.317199230194092,
|
|
"epoch": 2.4614472123368922,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004771357687917021,
|
|
"loss": 5.1759,
|
|
"mean_token_accuracy": 0.17290179580450057,
|
|
"num_tokens": 5960187.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.290809917449951,
|
|
"epoch": 2.4673784104389087,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004769261085063,
|
|
"loss": 5.1032,
|
|
"mean_token_accuracy": 0.17626044750213624,
|
|
"num_tokens": 5975211.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.2783082008361815,
|
|
"epoch": 2.473309608540925,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047671554327157996,
|
|
"loss": 5.09,
|
|
"mean_token_accuracy": 0.17717413753271102,
|
|
"num_tokens": 5988904.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.230610990524292,
|
|
"epoch": 2.4792408066429417,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047650407403120845,
|
|
"loss": 5.0964,
|
|
"mean_token_accuracy": 0.17791107147932053,
|
|
"num_tokens": 6004691.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.2898766040802006,
|
|
"epoch": 2.4851720047449586,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047629170173290296,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.17288617938756942,
|
|
"num_tokens": 6017859.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.320411825180054,
|
|
"epoch": 2.491103202846975,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004760784273284286,
|
|
"loss": 5.1645,
|
|
"mean_token_accuracy": 0.17397502213716506,
|
|
"num_tokens": 6032042.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.293785238265992,
|
|
"epoch": 2.4970344009489915,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004758642517735928,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.1742447093129158,
|
|
"num_tokens": 6048591.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.232989549636841,
|
|
"epoch": 2.5029655990510085,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047564917602824203,
|
|
"loss": 5.0759,
|
|
"mean_token_accuracy": 0.179649518430233,
|
|
"num_tokens": 6062436.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.266342687606811,
|
|
"epoch": 2.508896797153025,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004754332010562567,
|
|
"loss": 5.0365,
|
|
"mean_token_accuracy": 0.18351121246814728,
|
|
"num_tokens": 6075564.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.31279501914978,
|
|
"epoch": 2.5148279952550414,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004752163278255473,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.1722079709172249,
|
|
"num_tokens": 6088939.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.295576190948486,
|
|
"epoch": 2.5207591933570583,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004749985573080501,
|
|
"loss": 5.0956,
|
|
"mean_token_accuracy": 0.17826481014490128,
|
|
"num_tokens": 6101855.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.266389179229736,
|
|
"epoch": 2.526690391459075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004747798904797225,
|
|
"loss": 5.0738,
|
|
"mean_token_accuracy": 0.1764925792813301,
|
|
"num_tokens": 6116192.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.1693144798278805,
|
|
"epoch": 2.5326215895610913,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004745603283205387,
|
|
"loss": 5.0887,
|
|
"mean_token_accuracy": 0.17885565012693405,
|
|
"num_tokens": 6130741.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.278003215789795,
|
|
"epoch": 2.538552787663108,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047433987181448546,
|
|
"loss": 5.056,
|
|
"mean_token_accuracy": 0.180057655274868,
|
|
"num_tokens": 6144937.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.2536839008331295,
|
|
"epoch": 2.5444839857651247,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004741185219495579,
|
|
"loss": 5.1384,
|
|
"mean_token_accuracy": 0.1841862380504608,
|
|
"num_tokens": 6156747.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.199393939971924,
|
|
"epoch": 2.550415183867141,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004738962797177544,
|
|
"loss": 5.0336,
|
|
"mean_token_accuracy": 0.18659159690141677,
|
|
"num_tokens": 6169148.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.317109870910644,
|
|
"epoch": 2.5563463819691576,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047367314611507276,
|
|
"loss": 5.107,
|
|
"mean_token_accuracy": 0.17958347201347352,
|
|
"num_tokens": 6181881.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.259987688064575,
|
|
"epoch": 2.562277580071174,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047344912214150565,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.17894216775894164,
|
|
"num_tokens": 6196155.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.232990169525147,
|
|
"epoch": 2.568208778173191,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047322420880103566,
|
|
"loss": 5.124,
|
|
"mean_token_accuracy": 0.17405239790678023,
|
|
"num_tokens": 6210918.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.249849700927735,
|
|
"epoch": 2.5741399762752075,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047299840710163157,
|
|
"loss": 5.0786,
|
|
"mean_token_accuracy": 0.18167687505483626,
|
|
"num_tokens": 6225747.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.256479358673095,
|
|
"epoch": 2.580071174377224,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004727717180552431,
|
|
"loss": 5.137,
|
|
"mean_token_accuracy": 0.17358365505933762,
|
|
"num_tokens": 6239038.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.342339181900025,
|
|
"epoch": 2.586002372479241,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047254414267779686,
|
|
"loss": 5.2372,
|
|
"mean_token_accuracy": 0.17196187376976013,
|
|
"num_tokens": 6253346.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.303297472000122,
|
|
"epoch": 2.5919335705812574,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004723156819891917,
|
|
"loss": 5.2119,
|
|
"mean_token_accuracy": 0.17385910600423812,
|
|
"num_tokens": 6268060.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.266570806503296,
|
|
"epoch": 2.597864768683274,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004720863370132937,
|
|
"loss": 5.0388,
|
|
"mean_token_accuracy": 0.18387284874916077,
|
|
"num_tokens": 6283192.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.214200639724732,
|
|
"epoch": 2.603795966785291,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047185610877793245,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.18674758821725845,
|
|
"num_tokens": 6297511.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.277515506744384,
|
|
"epoch": 2.6097271648873073,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004716249983148955,
|
|
"loss": 5.1247,
|
|
"mean_token_accuracy": 0.1791581928730011,
|
|
"num_tokens": 6313257.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.248571062088013,
|
|
"epoch": 2.6156583629893237,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004713930066599246,
|
|
"loss": 5.0519,
|
|
"mean_token_accuracy": 0.1780620649456978,
|
|
"num_tokens": 6326437.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.23038158416748,
|
|
"epoch": 2.6215895610913407,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047116013485271014,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.18125442415475845,
|
|
"num_tokens": 6341704.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.240528726577759,
|
|
"epoch": 2.627520759193357,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004709263839368875,
|
|
"loss": 5.105,
|
|
"mean_token_accuracy": 0.17949420362710952,
|
|
"num_tokens": 6355783.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.3141090869903564,
|
|
"epoch": 2.6334519572953736,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047069175496003147,
|
|
"loss": 5.1015,
|
|
"mean_token_accuracy": 0.1741042762994766,
|
|
"num_tokens": 6370601.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.2879376888275145,
|
|
"epoch": 2.6393831553973905,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004704562489736522,
|
|
"loss": 5.1441,
|
|
"mean_token_accuracy": 0.1749232068657875,
|
|
"num_tokens": 6388124.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.316340398788452,
|
|
"epoch": 2.645314353499407,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004702198670331901,
|
|
"loss": 5.1487,
|
|
"mean_token_accuracy": 0.17196937203407286,
|
|
"num_tokens": 6403018.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.255807828903198,
|
|
"epoch": 2.6512455516014235,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004699826101980112,
|
|
"loss": 5.0898,
|
|
"mean_token_accuracy": 0.17568791508674622,
|
|
"num_tokens": 6417403.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.249996376037598,
|
|
"epoch": 2.65717674970344,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004697444795314027,
|
|
"loss": 5.1367,
|
|
"mean_token_accuracy": 0.1690463364124298,
|
|
"num_tokens": 6432950.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.27621431350708,
|
|
"epoch": 2.6631079478054565,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046950547610056766,
|
|
"loss": 5.1516,
|
|
"mean_token_accuracy": 0.1721343442797661,
|
|
"num_tokens": 6447735.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.281610584259033,
|
|
"epoch": 2.6690391459074734,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004692656009766206,
|
|
"loss": 5.0463,
|
|
"mean_token_accuracy": 0.18724312782287597,
|
|
"num_tokens": 6460066.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.2377824783325195,
|
|
"epoch": 2.67497034400949,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046902485523458266,
|
|
"loss": 5.1765,
|
|
"mean_token_accuracy": 0.16881112903356552,
|
|
"num_tokens": 6473880.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.330496311187744,
|
|
"epoch": 2.6809015421115063,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004687832399533767,
|
|
"loss": 5.0731,
|
|
"mean_token_accuracy": 0.1850060999393463,
|
|
"num_tokens": 6485678.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.215729856491089,
|
|
"epoch": 2.6868327402135233,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004685407562158226,
|
|
"loss": 5.1479,
|
|
"mean_token_accuracy": 0.17705656737089157,
|
|
"num_tokens": 6498606.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.218173503875732,
|
|
"epoch": 2.6927639383155397,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004682974051086321,
|
|
"loss": 5.0863,
|
|
"mean_token_accuracy": 0.18237122297286987,
|
|
"num_tokens": 6511590.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.29819803237915,
|
|
"epoch": 2.698695136417556,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004680531877224042,
|
|
"loss": 5.1659,
|
|
"mean_token_accuracy": 0.17502329647541046,
|
|
"num_tokens": 6525391.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.247089862823486,
|
|
"epoch": 2.704626334519573,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046780810515162027,
|
|
"loss": 5.1129,
|
|
"mean_token_accuracy": 0.17722094506025315,
|
|
"num_tokens": 6540749.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.228551435470581,
|
|
"epoch": 2.7105575326215896,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046756215849463905,
|
|
"loss": 5.1351,
|
|
"mean_token_accuracy": 0.1777515858411789,
|
|
"num_tokens": 6556607.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.227963972091675,
|
|
"epoch": 2.716488730723606,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046731534885369175,
|
|
"loss": 5.0378,
|
|
"mean_token_accuracy": 0.18670109808444976,
|
|
"num_tokens": 6569340.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.254456853866577,
|
|
"epoch": 2.722419928825623,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046706767733487714,
|
|
"loss": 5.1495,
|
|
"mean_token_accuracy": 0.168257737159729,
|
|
"num_tokens": 6586070.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.350936985015869,
|
|
"epoch": 2.7283511269276395,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046681914504815653,
|
|
"loss": 5.2384,
|
|
"mean_token_accuracy": 0.17244828790426253,
|
|
"num_tokens": 6601441.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.22352089881897,
|
|
"epoch": 2.734282325029656,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004665697531073489,
|
|
"loss": 5.1004,
|
|
"mean_token_accuracy": 0.17834114283323288,
|
|
"num_tokens": 6616478.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.29279670715332,
|
|
"epoch": 2.7402135231316724,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004663195026301258,
|
|
"loss": 5.1774,
|
|
"mean_token_accuracy": 0.17028742134571076,
|
|
"num_tokens": 6631519.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.289520025253296,
|
|
"epoch": 2.7461447212336894,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046606839473800626,
|
|
"loss": 5.1519,
|
|
"mean_token_accuracy": 0.1711852639913559,
|
|
"num_tokens": 6646862.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.285637235641479,
|
|
"epoch": 2.752075919335706,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046581643055635215,
|
|
"loss": 5.0706,
|
|
"mean_token_accuracy": 0.18027766644954682,
|
|
"num_tokens": 6660731.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.257721757888794,
|
|
"epoch": 2.7580071174377223,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046556361121436265,
|
|
"loss": 5.0884,
|
|
"mean_token_accuracy": 0.17217349857091904,
|
|
"num_tokens": 6675922.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.210122585296631,
|
|
"epoch": 2.763938315539739,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004653099378450696,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.17834757566452025,
|
|
"num_tokens": 6689363.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.289632892608642,
|
|
"epoch": 2.7698695136417557,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004650554115853321,
|
|
"loss": 5.08,
|
|
"mean_token_accuracy": 0.17782238125801086,
|
|
"num_tokens": 6702655.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.233364868164062,
|
|
"epoch": 2.775800711743772,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046480003357583154,
|
|
"loss": 5.115,
|
|
"mean_token_accuracy": 0.1791711837053299,
|
|
"num_tokens": 6718101.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.249103927612305,
|
|
"epoch": 2.7817319098457887,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046454380496106665,
|
|
"loss": 5.0327,
|
|
"mean_token_accuracy": 0.18523623198270797,
|
|
"num_tokens": 6731410.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.2448647022247314,
|
|
"epoch": 2.7876631079478056,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004642867268893481,
|
|
"loss": 5.1059,
|
|
"mean_token_accuracy": 0.18096584379673003,
|
|
"num_tokens": 6744544.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.1900553703308105,
|
|
"epoch": 2.793594306049822,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004640288005127933,
|
|
"loss": 5.0096,
|
|
"mean_token_accuracy": 0.18570955991744995,
|
|
"num_tokens": 6757559.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.22004246711731,
|
|
"epoch": 2.7995255041518385,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000463770026987322,
|
|
"loss": 5.0821,
|
|
"mean_token_accuracy": 0.17724673599004745,
|
|
"num_tokens": 6772655.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.193439579010009,
|
|
"epoch": 2.8054567022538555,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004635104074726498,
|
|
"loss": 5.011,
|
|
"mean_token_accuracy": 0.18544310331344604,
|
|
"num_tokens": 6786038.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.228322505950928,
|
|
"epoch": 2.811387900355872,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004632499431322842,
|
|
"loss": 5.0586,
|
|
"mean_token_accuracy": 0.17480345368385314,
|
|
"num_tokens": 6799759.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.192971563339233,
|
|
"epoch": 2.8173190984578884,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046298863513351854,
|
|
"loss": 5.1025,
|
|
"mean_token_accuracy": 0.1785614773631096,
|
|
"num_tokens": 6814242.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.30884952545166,
|
|
"epoch": 2.8232502965599053,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00046272648464742733,
|
|
"loss": 5.1246,
|
|
"mean_token_accuracy": 0.16953468918800355,
|
|
"num_tokens": 6828697.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.15752215385437,
|
|
"epoch": 2.829181494661922,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004624634928488605,
|
|
"loss": 5.0758,
|
|
"mean_token_accuracy": 0.18574994206428527,
|
|
"num_tokens": 6842549.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.288483715057373,
|
|
"epoch": 2.8351126927639383,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046219966091643865,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.17601943910121917,
|
|
"num_tokens": 6857599.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.216753959655762,
|
|
"epoch": 2.8410438908659548,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046193499003254737,
|
|
"loss": 5.0904,
|
|
"mean_token_accuracy": 0.18286330252885818,
|
|
"num_tokens": 6872621.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.251305961608887,
|
|
"epoch": 2.8469750889679717,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004616694813833322,
|
|
"loss": 5.1589,
|
|
"mean_token_accuracy": 0.1729411706328392,
|
|
"num_tokens": 6887297.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.263056373596191,
|
|
"epoch": 2.852906287069988,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046140313615869287,
|
|
"loss": 5.0538,
|
|
"mean_token_accuracy": 0.18049194067716598,
|
|
"num_tokens": 6900857.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.1703661441802975,
|
|
"epoch": 2.8588374851720046,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046113595555227885,
|
|
"loss": 5.0715,
|
|
"mean_token_accuracy": 0.18290164321660995,
|
|
"num_tokens": 6917351.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.134192991256714,
|
|
"epoch": 2.864768683274021,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000460867940761483,
|
|
"loss": 5.0086,
|
|
"mean_token_accuracy": 0.19504904747009277,
|
|
"num_tokens": 6930619.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.185745000839233,
|
|
"epoch": 2.870699881376038,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004605990929874366,
|
|
"loss": 5.0501,
|
|
"mean_token_accuracy": 0.17975494414567947,
|
|
"num_tokens": 6944842.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.221125698089599,
|
|
"epoch": 2.8766310794780545,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046032941343500443,
|
|
"loss": 5.0331,
|
|
"mean_token_accuracy": 0.18169520050287247,
|
|
"num_tokens": 6957699.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.243278074264526,
|
|
"epoch": 2.882562277580071,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046005890331277875,
|
|
"loss": 5.1779,
|
|
"mean_token_accuracy": 0.1789306804537773,
|
|
"num_tokens": 6973373.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.273877429962158,
|
|
"epoch": 2.888493475682088,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004597875638330739,
|
|
"loss": 5.1065,
|
|
"mean_token_accuracy": 0.17897169589996337,
|
|
"num_tokens": 6985943.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.267488431930542,
|
|
"epoch": 2.8944246737841044,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045951539621192137,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.19029628336429597,
|
|
"num_tokens": 6999655.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.225543117523193,
|
|
"epoch": 2.900355871886121,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004592424016690637,
|
|
"loss": 5.0894,
|
|
"mean_token_accuracy": 0.1805782213807106,
|
|
"num_tokens": 7014278.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.211720752716064,
|
|
"epoch": 2.906287069988138,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045896858142794976,
|
|
"loss": 5.0853,
|
|
"mean_token_accuracy": 0.18050264269113542,
|
|
"num_tokens": 7029998.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.261847639083863,
|
|
"epoch": 2.9122182680901543,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004586939367157285,
|
|
"loss": 5.08,
|
|
"mean_token_accuracy": 0.17528700977563857,
|
|
"num_tokens": 7043212.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.205530023574829,
|
|
"epoch": 2.9181494661921707,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000458418468763244,
|
|
"loss": 5.0157,
|
|
"mean_token_accuracy": 0.18455137461423873,
|
|
"num_tokens": 7057723.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.222577953338623,
|
|
"epoch": 2.9240806642941877,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045814217880502973,
|
|
"loss": 5.1525,
|
|
"mean_token_accuracy": 0.17377886772155762,
|
|
"num_tokens": 7073334.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.240353107452393,
|
|
"epoch": 2.930011862396204,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000457865068079303,
|
|
"loss": 5.1136,
|
|
"mean_token_accuracy": 0.17955139875411988,
|
|
"num_tokens": 7087671.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.162194871902466,
|
|
"epoch": 2.9359430604982206,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045758713782795946,
|
|
"loss": 4.9601,
|
|
"mean_token_accuracy": 0.18688122928142548,
|
|
"num_tokens": 7101168.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.209585046768188,
|
|
"epoch": 2.941874258600237,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004573083892965677,
|
|
"loss": 5.1268,
|
|
"mean_token_accuracy": 0.18048435896635057,
|
|
"num_tokens": 7115581.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.200497722625732,
|
|
"epoch": 2.9478054567022536,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045702882373436317,
|
|
"loss": 5.1,
|
|
"mean_token_accuracy": 0.17545540034770965,
|
|
"num_tokens": 7129948.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.13970685005188,
|
|
"epoch": 2.9537366548042705,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004567484423942433,
|
|
"loss": 4.9933,
|
|
"mean_token_accuracy": 0.18735943138599395,
|
|
"num_tokens": 7144122.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.212256240844726,
|
|
"epoch": 2.959667852906287,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045646724653276117,
|
|
"loss": 5.0423,
|
|
"mean_token_accuracy": 0.19172451496124268,
|
|
"num_tokens": 7157892.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.22672963142395,
|
|
"epoch": 2.9655990510083035,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045618523741012044,
|
|
"loss": 5.1198,
|
|
"mean_token_accuracy": 0.1785096600651741,
|
|
"num_tokens": 7172581.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.9655990510083035,
|
|
"eval_entropy": 5.128191098113732,
|
|
"eval_loss": 5.4554667472839355,
|
|
"eval_mean_token_accuracy": 0.16967149755340413,
|
|
"eval_num_tokens": 7172581.0,
|
|
"eval_runtime": 2.4682,
|
|
"eval_samples_per_second": 1056.643,
|
|
"eval_steps_per_second": 132.08,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.209706211090088,
|
|
"epoch": 2.9715302491103204,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045590241629016933,
|
|
"loss": 5.081,
|
|
"mean_token_accuracy": 0.17838001102209092,
|
|
"num_tokens": 7187180.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.186795330047607,
|
|
"epoch": 2.977461447212337,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045561878444039526,
|
|
"loss": 5.0928,
|
|
"mean_token_accuracy": 0.1853885158896446,
|
|
"num_tokens": 7202153.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.233902025222778,
|
|
"epoch": 2.9833926453143533,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045533434313191866,
|
|
"loss": 5.0663,
|
|
"mean_token_accuracy": 0.1795577213168144,
|
|
"num_tokens": 7217714.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.072184324264526,
|
|
"epoch": 2.9893238434163703,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004550490936394881,
|
|
"loss": 5.0248,
|
|
"mean_token_accuracy": 0.1885547786951065,
|
|
"num_tokens": 7232366.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.288804340362549,
|
|
"epoch": 2.9952550415183867,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004547630372414737,
|
|
"loss": 5.0433,
|
|
"mean_token_accuracy": 0.18170098960399628,
|
|
"num_tokens": 7245542.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.13629674911499,
|
|
"epoch": 3.001186239620403,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045447617521986205,
|
|
"loss": 4.966,
|
|
"mean_token_accuracy": 0.18382924646139145,
|
|
"num_tokens": 7260437.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.186529493331909,
|
|
"epoch": 3.00711743772242,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045418850886025006,
|
|
"loss": 4.7908,
|
|
"mean_token_accuracy": 0.19730883985757827,
|
|
"num_tokens": 7273470.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.14306674003601,
|
|
"epoch": 3.0130486358244366,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004539000394518393,
|
|
"loss": 4.8184,
|
|
"mean_token_accuracy": 0.19704275578260422,
|
|
"num_tokens": 7286515.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.08558144569397,
|
|
"epoch": 3.018979833926453,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045361076828743054,
|
|
"loss": 4.6494,
|
|
"mean_token_accuracy": 0.2086326450109482,
|
|
"num_tokens": 7299011.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.100022411346435,
|
|
"epoch": 3.0249110320284696,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045332069666341733,
|
|
"loss": 4.7556,
|
|
"mean_token_accuracy": 0.2030994936823845,
|
|
"num_tokens": 7312421.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.228758382797241,
|
|
"epoch": 3.0308422301304865,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000453029825879781,
|
|
"loss": 4.829,
|
|
"mean_token_accuracy": 0.1881020963191986,
|
|
"num_tokens": 7326298.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.20881233215332,
|
|
"epoch": 3.036773428232503,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004527381572400838,
|
|
"loss": 4.8349,
|
|
"mean_token_accuracy": 0.19809474796056747,
|
|
"num_tokens": 7340411.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.2084184169769285,
|
|
"epoch": 3.0427046263345194,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045244569205146425,
|
|
"loss": 4.7739,
|
|
"mean_token_accuracy": 0.19972959756851197,
|
|
"num_tokens": 7355853.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.0268631935119625,
|
|
"epoch": 3.0486358244365364,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045215243162463023,
|
|
"loss": 4.7546,
|
|
"mean_token_accuracy": 0.2027596727013588,
|
|
"num_tokens": 7370462.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.1120353698730465,
|
|
"epoch": 3.054567022538553,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045185837727385373,
|
|
"loss": 4.7613,
|
|
"mean_token_accuracy": 0.19956568926572799,
|
|
"num_tokens": 7382283.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.185872745513916,
|
|
"epoch": 3.0604982206405693,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004515635303169648,
|
|
"loss": 4.8004,
|
|
"mean_token_accuracy": 0.19398040175437928,
|
|
"num_tokens": 7397282.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.129135847091675,
|
|
"epoch": 3.066429418742586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045126789207534544,
|
|
"loss": 4.7633,
|
|
"mean_token_accuracy": 0.19493884593248367,
|
|
"num_tokens": 7410471.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.097009515762329,
|
|
"epoch": 3.0723606168446027,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045097146387392423,
|
|
"loss": 4.7541,
|
|
"mean_token_accuracy": 0.20093791782855988,
|
|
"num_tokens": 7426151.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.208783435821533,
|
|
"epoch": 3.078291814946619,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045067424704116956,
|
|
"loss": 4.8063,
|
|
"mean_token_accuracy": 0.19725323617458343,
|
|
"num_tokens": 7441336.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.1333612442016605,
|
|
"epoch": 3.0842230130486357,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004503762429090845,
|
|
"loss": 4.8512,
|
|
"mean_token_accuracy": 0.19190199822187423,
|
|
"num_tokens": 7456152.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.071452474594116,
|
|
"epoch": 3.0901542111506526,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045007745281320035,
|
|
"loss": 4.7492,
|
|
"mean_token_accuracy": 0.19597499817609787,
|
|
"num_tokens": 7469477.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.152786159515381,
|
|
"epoch": 3.096085409252669,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004497778780925706,
|
|
"loss": 4.745,
|
|
"mean_token_accuracy": 0.20036472231149674,
|
|
"num_tokens": 7482859.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.149897813796997,
|
|
"epoch": 3.1020166073546855,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044947752008976546,
|
|
"loss": 4.7565,
|
|
"mean_token_accuracy": 0.20434682816267014,
|
|
"num_tokens": 7497021.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.082270812988281,
|
|
"epoch": 3.1079478054567025,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004491763801508652,
|
|
"loss": 4.8287,
|
|
"mean_token_accuracy": 0.19397101253271104,
|
|
"num_tokens": 7512528.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.1444420337677,
|
|
"epoch": 3.113879003558719,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004488744596254545,
|
|
"loss": 4.8243,
|
|
"mean_token_accuracy": 0.19616906642913817,
|
|
"num_tokens": 7529492.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.099391889572144,
|
|
"epoch": 3.1198102016607354,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044857175986661634,
|
|
"loss": 4.8307,
|
|
"mean_token_accuracy": 0.1908296301960945,
|
|
"num_tokens": 7544750.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.151020145416259,
|
|
"epoch": 3.125741399762752,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044826828223092584,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.19755022823810578,
|
|
"num_tokens": 7560436.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.078480672836304,
|
|
"epoch": 3.131672597864769,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004479640280784443,
|
|
"loss": 4.744,
|
|
"mean_token_accuracy": 0.19955695867538453,
|
|
"num_tokens": 7572784.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.113914966583252,
|
|
"epoch": 3.1376037959667853,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044765899877271306,
|
|
"loss": 4.859,
|
|
"mean_token_accuracy": 0.1986508622765541,
|
|
"num_tokens": 7585677.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.126020288467407,
|
|
"epoch": 3.1435349940688018,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004473531956807471,
|
|
"loss": 4.8702,
|
|
"mean_token_accuracy": 0.19153172969818116,
|
|
"num_tokens": 7600099.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.072196006774902,
|
|
"epoch": 3.1494661921708187,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004470466201730298,
|
|
"loss": 4.7925,
|
|
"mean_token_accuracy": 0.19439838826656342,
|
|
"num_tokens": 7615907.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.143037557601929,
|
|
"epoch": 3.155397390272835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044673927362350545,
|
|
"loss": 4.8237,
|
|
"mean_token_accuracy": 0.19680167287588118,
|
|
"num_tokens": 7629322.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.135795927047729,
|
|
"epoch": 3.1613285883748516,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044643115740957437,
|
|
"loss": 4.7916,
|
|
"mean_token_accuracy": 0.1951526179909706,
|
|
"num_tokens": 7642876.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.147537422180176,
|
|
"epoch": 3.167259786476868,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000446122272912086,
|
|
"loss": 4.7397,
|
|
"mean_token_accuracy": 0.20763775259256362,
|
|
"num_tokens": 7655252.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.109446811676025,
|
|
"epoch": 3.173190984578885,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004458126215153329,
|
|
"loss": 4.7983,
|
|
"mean_token_accuracy": 0.19524380266666413,
|
|
"num_tokens": 7669858.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.115171337127686,
|
|
"epoch": 3.1791221826809015,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004455022046070446,
|
|
"loss": 4.8484,
|
|
"mean_token_accuracy": 0.18973655700683595,
|
|
"num_tokens": 7684268.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.018248653411865,
|
|
"epoch": 3.185053380782918,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004451910235783814,
|
|
"loss": 4.717,
|
|
"mean_token_accuracy": 0.21085696369409562,
|
|
"num_tokens": 7696579.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.127526092529297,
|
|
"epoch": 3.190984578884935,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004448790798239278,
|
|
"loss": 4.8436,
|
|
"mean_token_accuracy": 0.19351662397384645,
|
|
"num_tokens": 7712167.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.105394887924194,
|
|
"epoch": 3.1969157769869514,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004445663747416868,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.19317025691270828,
|
|
"num_tokens": 7729466.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.043076610565185,
|
|
"epoch": 3.202846975088968,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044425290973307345,
|
|
"loss": 4.6998,
|
|
"mean_token_accuracy": 0.20704733729362487,
|
|
"num_tokens": 7741327.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.05340690612793,
|
|
"epoch": 3.208778173190985,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004439386862029081,
|
|
"loss": 4.7979,
|
|
"mean_token_accuracy": 0.2003027930855751,
|
|
"num_tokens": 7757212.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.1013377666473385,
|
|
"epoch": 3.2147093712930013,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044362370555941057,
|
|
"loss": 4.8373,
|
|
"mean_token_accuracy": 0.19453289955854416,
|
|
"num_tokens": 7772726.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.200697374343872,
|
|
"epoch": 3.2206405693950177,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004433079692141941,
|
|
"loss": 4.9332,
|
|
"mean_token_accuracy": 0.1931446298956871,
|
|
"num_tokens": 7788185.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.098084259033203,
|
|
"epoch": 3.2265717674970342,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044299147858225835,
|
|
"loss": 4.8124,
|
|
"mean_token_accuracy": 0.19363399744033813,
|
|
"num_tokens": 7802653.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.1125976085662845,
|
|
"epoch": 3.232502965599051,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004426742350819834,
|
|
"loss": 4.7917,
|
|
"mean_token_accuracy": 0.19389940202236175,
|
|
"num_tokens": 7816898.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.147168445587158,
|
|
"epoch": 3.2384341637010676,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004423562401351234,
|
|
"loss": 4.876,
|
|
"mean_token_accuracy": 0.1905128002166748,
|
|
"num_tokens": 7833509.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.192942333221436,
|
|
"epoch": 3.244365361803084,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044203749516680035,
|
|
"loss": 4.8427,
|
|
"mean_token_accuracy": 0.1987154722213745,
|
|
"num_tokens": 7849910.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.063848209381104,
|
|
"epoch": 3.250296559905101,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004417180016054972,
|
|
"loss": 4.8113,
|
|
"mean_token_accuracy": 0.1946829855442047,
|
|
"num_tokens": 7864555.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.090720558166504,
|
|
"epoch": 3.2562277580071175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044139776088305215,
|
|
"loss": 4.7492,
|
|
"mean_token_accuracy": 0.2016763925552368,
|
|
"num_tokens": 7878850.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.091657066345215,
|
|
"epoch": 3.262158956109134,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044107677443465165,
|
|
"loss": 4.9031,
|
|
"mean_token_accuracy": 0.19536128789186477,
|
|
"num_tokens": 7894200.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.134050750732422,
|
|
"epoch": 3.2680901542111505,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044075504369882416,
|
|
"loss": 4.8529,
|
|
"mean_token_accuracy": 0.19184636920690537,
|
|
"num_tokens": 7908939.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.132843923568726,
|
|
"epoch": 3.2740213523131674,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004404325701174338,
|
|
"loss": 4.8618,
|
|
"mean_token_accuracy": 0.19030990153551103,
|
|
"num_tokens": 7924223.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.081524896621704,
|
|
"epoch": 3.279952550415184,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004401093551356739,
|
|
"loss": 4.7557,
|
|
"mean_token_accuracy": 0.2024812862277031,
|
|
"num_tokens": 7938323.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.049100589752197,
|
|
"epoch": 3.2858837485172003,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004397854002020604,
|
|
"loss": 4.7814,
|
|
"mean_token_accuracy": 0.2020518437027931,
|
|
"num_tokens": 7952126.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.080816555023193,
|
|
"epoch": 3.2918149466192173,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004394607067684251,
|
|
"loss": 4.7988,
|
|
"mean_token_accuracy": 0.19751653522253038,
|
|
"num_tokens": 7965150.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.15736985206604,
|
|
"epoch": 3.2977461447212337,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043913527628990984,
|
|
"loss": 4.8365,
|
|
"mean_token_accuracy": 0.19114754498004913,
|
|
"num_tokens": 7980532.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.0498096466064455,
|
|
"epoch": 3.30367734282325,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043880911022495955,
|
|
"loss": 4.7979,
|
|
"mean_token_accuracy": 0.19996661990880965,
|
|
"num_tokens": 7994731.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.045728778839111,
|
|
"epoch": 3.309608540925267,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004384822100353155,
|
|
"loss": 4.7792,
|
|
"mean_token_accuracy": 0.20297738313674926,
|
|
"num_tokens": 8009405.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.043602848052979,
|
|
"epoch": 3.3155397390272836,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043815457718600933,
|
|
"loss": 4.703,
|
|
"mean_token_accuracy": 0.20316480547189714,
|
|
"num_tokens": 8023683.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 4.979565191268921,
|
|
"epoch": 3.3214709371293,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004378262131453558,
|
|
"loss": 4.7448,
|
|
"mean_token_accuracy": 0.20116145312786102,
|
|
"num_tokens": 8038219.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.029000902175904,
|
|
"epoch": 3.3274021352313166,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043749711938494693,
|
|
"loss": 4.7479,
|
|
"mean_token_accuracy": 0.2009764328598976,
|
|
"num_tokens": 8052037.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.063984489440918,
|
|
"epoch": 3.3333333333333335,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043716729737964475,
|
|
"loss": 4.8005,
|
|
"mean_token_accuracy": 0.19967271238565446,
|
|
"num_tokens": 8065346.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.089534330368042,
|
|
"epoch": 3.33926453143535,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000436836748607575,
|
|
"loss": 4.8024,
|
|
"mean_token_accuracy": 0.19638994485139846,
|
|
"num_tokens": 8081411.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.120129728317261,
|
|
"epoch": 3.3451957295373664,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004365054745501209,
|
|
"loss": 4.8472,
|
|
"mean_token_accuracy": 0.19565204232931138,
|
|
"num_tokens": 8096979.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.020432186126709,
|
|
"epoch": 3.351126927639383,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004361734766919156,
|
|
"loss": 4.8349,
|
|
"mean_token_accuracy": 0.1926871970295906,
|
|
"num_tokens": 8110573.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.061763572692871,
|
|
"epoch": 3.3570581257414,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004358407565208361,
|
|
"loss": 4.8833,
|
|
"mean_token_accuracy": 0.1976574257016182,
|
|
"num_tokens": 8125529.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.12091588973999,
|
|
"epoch": 3.3629893238434163,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00043550731552799676,
|
|
"loss": 4.8466,
|
|
"mean_token_accuracy": 0.19565797746181487,
|
|
"num_tokens": 8141042.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.022051239013672,
|
|
"epoch": 3.368920521945433,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004351731552077423,
|
|
"loss": 4.7575,
|
|
"mean_token_accuracy": 0.19804837554693222,
|
|
"num_tokens": 8153157.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.021864032745361,
|
|
"epoch": 3.3748517200474497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043483827705764094,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.20032886266708375,
|
|
"num_tokens": 8168570.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.108405017852784,
|
|
"epoch": 3.380782918149466,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043450268257847814,
|
|
"loss": 4.8059,
|
|
"mean_token_accuracy": 0.19953808933496475,
|
|
"num_tokens": 8182041.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.110202646255493,
|
|
"epoch": 3.3867141162514827,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043416637327424955,
|
|
"loss": 4.8449,
|
|
"mean_token_accuracy": 0.19766062647104263,
|
|
"num_tokens": 8196466.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.049158334732056,
|
|
"epoch": 3.3926453143534996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043382935065215444,
|
|
"loss": 4.7834,
|
|
"mean_token_accuracy": 0.19956779778003692,
|
|
"num_tokens": 8210847.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.013025093078613,
|
|
"epoch": 3.398576512455516,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004334916162225889,
|
|
"loss": 4.7041,
|
|
"mean_token_accuracy": 0.2068425863981247,
|
|
"num_tokens": 8225575.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.076068353652954,
|
|
"epoch": 3.4045077105575325,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043315317149913876,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.19905340373516084,
|
|
"num_tokens": 8239881.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.018644142150879,
|
|
"epoch": 3.4104389086595495,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043281401799857356,
|
|
"loss": 4.8426,
|
|
"mean_token_accuracy": 0.19684257358312607,
|
|
"num_tokens": 8253692.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.0646027565002445,
|
|
"epoch": 3.416370106761566,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043247415724083886,
|
|
"loss": 4.7819,
|
|
"mean_token_accuracy": 0.19912485182285308,
|
|
"num_tokens": 8268322.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.025012445449829,
|
|
"epoch": 3.4223013048635824,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004321335907490502,
|
|
"loss": 4.8053,
|
|
"mean_token_accuracy": 0.19429467767477035,
|
|
"num_tokens": 8285317.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.100164270401001,
|
|
"epoch": 3.428232502965599,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004317923200494856,
|
|
"loss": 4.7933,
|
|
"mean_token_accuracy": 0.20158445090055466,
|
|
"num_tokens": 8299429.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.039232349395752,
|
|
"epoch": 3.434163701067616,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043145034667157935,
|
|
"loss": 4.8165,
|
|
"mean_token_accuracy": 0.1936866894364357,
|
|
"num_tokens": 8314039.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.04435977935791,
|
|
"epoch": 3.4400948991696323,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043110767214791466,
|
|
"loss": 4.8579,
|
|
"mean_token_accuracy": 0.19030549228191376,
|
|
"num_tokens": 8328634.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.061645078659057,
|
|
"epoch": 3.4460260972716488,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004307642980142171,
|
|
"loss": 4.7607,
|
|
"mean_token_accuracy": 0.20237151086330413,
|
|
"num_tokens": 8342448.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.0385942459106445,
|
|
"epoch": 3.4519572953736652,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004304202258093473,
|
|
"loss": 4.7632,
|
|
"mean_token_accuracy": 0.20581259578466415,
|
|
"num_tokens": 8354450.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.031369161605835,
|
|
"epoch": 3.457888493475682,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043007545707529496,
|
|
"loss": 4.8276,
|
|
"mean_token_accuracy": 0.20016095489263536,
|
|
"num_tokens": 8369054.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.084920263290405,
|
|
"epoch": 3.4638196915776986,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004297299933571708,
|
|
"loss": 4.8296,
|
|
"mean_token_accuracy": 0.20116593837738037,
|
|
"num_tokens": 8383401.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.0826586246490475,
|
|
"epoch": 3.469750889679715,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004293838362032003,
|
|
"loss": 4.8431,
|
|
"mean_token_accuracy": 0.19186071306467056,
|
|
"num_tokens": 8396990.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.091199779510498,
|
|
"epoch": 3.475682087781732,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042903698716471677,
|
|
"loss": 4.7678,
|
|
"mean_token_accuracy": 0.2013833075761795,
|
|
"num_tokens": 8411212.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.048702621459961,
|
|
"epoch": 3.4816132858837485,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004286894477961541,
|
|
"loss": 4.816,
|
|
"mean_token_accuracy": 0.19629474729299545,
|
|
"num_tokens": 8424669.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 4.976538515090942,
|
|
"epoch": 3.487544483985765,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042834121965504014,
|
|
"loss": 4.7754,
|
|
"mean_token_accuracy": 0.19990172535181044,
|
|
"num_tokens": 8437122.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.10742974281311,
|
|
"epoch": 3.493475682087782,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00042799230430198935,
|
|
"loss": 4.8296,
|
|
"mean_token_accuracy": 0.1914140284061432,
|
|
"num_tokens": 8450543.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.064130353927612,
|
|
"epoch": 3.4994068801897984,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000427642703300696,
|
|
"loss": 4.7662,
|
|
"mean_token_accuracy": 0.2023661881685257,
|
|
"num_tokens": 8464720.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.0957324504852295,
|
|
"epoch": 3.505338078291815,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042729241821792726,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.1962006002664566,
|
|
"num_tokens": 8478547.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.1516471862792965,
|
|
"epoch": 3.511269276393832,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000426941450623516,
|
|
"loss": 4.8781,
|
|
"mean_token_accuracy": 0.19749569445848464,
|
|
"num_tokens": 8494126.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.031353759765625,
|
|
"epoch": 3.5172004744958483,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042658980209035394,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.186987940967083,
|
|
"num_tokens": 8508575.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.0075500965118405,
|
|
"epoch": 3.5231316725978647,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042623747419438427,
|
|
"loss": 4.7508,
|
|
"mean_token_accuracy": 0.1991992712020874,
|
|
"num_tokens": 8523589.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.158101797103882,
|
|
"epoch": 3.5290628706998812,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000425884468514595,
|
|
"loss": 4.8535,
|
|
"mean_token_accuracy": 0.19646587669849397,
|
|
"num_tokens": 8538238.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 4.996473217010498,
|
|
"epoch": 3.534994068801898,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042553078663301163,
|
|
"loss": 4.8191,
|
|
"mean_token_accuracy": 0.20561834871768953,
|
|
"num_tokens": 8551959.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.088233804702758,
|
|
"epoch": 3.5409252669039146,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042517643013469,
|
|
"loss": 4.8048,
|
|
"mean_token_accuracy": 0.20215945243835448,
|
|
"num_tokens": 8564628.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.071303081512451,
|
|
"epoch": 3.546856465005931,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004248214006077095,
|
|
"loss": 4.821,
|
|
"mean_token_accuracy": 0.19784992188215256,
|
|
"num_tokens": 8579342.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.061466932296753,
|
|
"epoch": 3.5527876631079476,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042446569964316557,
|
|
"loss": 4.8055,
|
|
"mean_token_accuracy": 0.19897904992103577,
|
|
"num_tokens": 8595096.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.215964984893799,
|
|
"epoch": 3.5587188612099645,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042410932883516297,
|
|
"loss": 4.9033,
|
|
"mean_token_accuracy": 0.1953563556075096,
|
|
"num_tokens": 8608023.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.5587188612099645,
|
|
"eval_entropy": 4.9771839255935575,
|
|
"eval_loss": 5.412055492401123,
|
|
"eval_mean_token_accuracy": 0.17675103251736587,
|
|
"eval_num_tokens": 8608023.0,
|
|
"eval_runtime": 2.4284,
|
|
"eval_samples_per_second": 1073.944,
|
|
"eval_steps_per_second": 134.243,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.042902183532715,
|
|
"epoch": 3.564650059311981,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042375228978080813,
|
|
"loss": 4.7649,
|
|
"mean_token_accuracy": 0.2008129134774208,
|
|
"num_tokens": 8623529.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.038916540145874,
|
|
"epoch": 3.5705812574139975,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042339458408020267,
|
|
"loss": 4.8383,
|
|
"mean_token_accuracy": 0.19716774225234984,
|
|
"num_tokens": 8639007.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.000910806655884,
|
|
"epoch": 3.5765124555160144,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004230362133364354,
|
|
"loss": 4.7451,
|
|
"mean_token_accuracy": 0.20509970784187317,
|
|
"num_tokens": 8653678.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.084241008758545,
|
|
"epoch": 3.582443653618031,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004226771791555758,
|
|
"loss": 4.9001,
|
|
"mean_token_accuracy": 0.18916919380426406,
|
|
"num_tokens": 8668663.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.098063039779663,
|
|
"epoch": 3.5883748517200473,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042231748314666684,
|
|
"loss": 4.8285,
|
|
"mean_token_accuracy": 0.1972780004143715,
|
|
"num_tokens": 8682121.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 4.962451457977295,
|
|
"epoch": 3.5943060498220643,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042195712692171705,
|
|
"loss": 4.8254,
|
|
"mean_token_accuracy": 0.19526612013578415,
|
|
"num_tokens": 8698409.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.078032970428467,
|
|
"epoch": 3.6002372479240807,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042159611209569416,
|
|
"loss": 4.8473,
|
|
"mean_token_accuracy": 0.1929122567176819,
|
|
"num_tokens": 8714076.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.09345006942749,
|
|
"epoch": 3.606168446026097,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042123444028651725,
|
|
"loss": 4.7946,
|
|
"mean_token_accuracy": 0.20138713866472244,
|
|
"num_tokens": 8730482.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 4.995736932754516,
|
|
"epoch": 3.612099644128114,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00042087211311504986,
|
|
"loss": 4.7808,
|
|
"mean_token_accuracy": 0.2021682783961296,
|
|
"num_tokens": 8744036.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.092488193511963,
|
|
"epoch": 3.6180308422301306,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042050913220509253,
|
|
"loss": 4.8339,
|
|
"mean_token_accuracy": 0.1878429561853409,
|
|
"num_tokens": 8760458.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.027280521392822,
|
|
"epoch": 3.623962040332147,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042014549918337554,
|
|
"loss": 4.788,
|
|
"mean_token_accuracy": 0.19931856244802476,
|
|
"num_tokens": 8774946.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.0445245742797855,
|
|
"epoch": 3.6298932384341636,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004197812156795518,
|
|
"loss": 4.7885,
|
|
"mean_token_accuracy": 0.20058033168315886,
|
|
"num_tokens": 8790918.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.032378005981445,
|
|
"epoch": 3.63582443653618,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004194162833261893,
|
|
"loss": 4.8178,
|
|
"mean_token_accuracy": 0.1946185812354088,
|
|
"num_tokens": 8803505.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.082729768753052,
|
|
"epoch": 3.641755634638197,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004190507037587639,
|
|
"loss": 4.8787,
|
|
"mean_token_accuracy": 0.1933801293373108,
|
|
"num_tokens": 8819567.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.109076070785522,
|
|
"epoch": 3.6476868327402134,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004186844786156522,
|
|
"loss": 4.7838,
|
|
"mean_token_accuracy": 0.18985679149627685,
|
|
"num_tokens": 8834257.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.0223876476287845,
|
|
"epoch": 3.65361803084223,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004183176095381238,
|
|
"loss": 4.7938,
|
|
"mean_token_accuracy": 0.2065998211503029,
|
|
"num_tokens": 8847866.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.010385465621948,
|
|
"epoch": 3.659549228944247,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000417950098170334,
|
|
"loss": 4.8118,
|
|
"mean_token_accuracy": 0.19870606809854507,
|
|
"num_tokens": 8863297.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.062593126296997,
|
|
"epoch": 3.6654804270462633,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000417581946159317,
|
|
"loss": 4.817,
|
|
"mean_token_accuracy": 0.19981457144021988,
|
|
"num_tokens": 8876605.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 4.998176097869873,
|
|
"epoch": 3.67141162514828,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004172131551549778,
|
|
"loss": 4.7572,
|
|
"mean_token_accuracy": 0.206072661280632,
|
|
"num_tokens": 8892040.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 4.948314142227173,
|
|
"epoch": 3.6773428232502967,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00041684372681008506,
|
|
"loss": 4.8146,
|
|
"mean_token_accuracy": 0.19830800145864486,
|
|
"num_tokens": 8905806.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.009223079681396,
|
|
"epoch": 3.683274021352313,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041647366278026415,
|
|
"loss": 4.777,
|
|
"mean_token_accuracy": 0.2011968970298767,
|
|
"num_tokens": 8920638.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.031396293640137,
|
|
"epoch": 3.6892052194543297,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041610296472398875,
|
|
"loss": 4.7595,
|
|
"mean_token_accuracy": 0.1939758390188217,
|
|
"num_tokens": 8933501.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.029152154922485,
|
|
"epoch": 3.6951364175563466,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041573163430257435,
|
|
"loss": 4.8538,
|
|
"mean_token_accuracy": 0.19719393253326417,
|
|
"num_tokens": 8948909.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.086005306243896,
|
|
"epoch": 3.701067615658363,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041535967318017037,
|
|
"loss": 4.8204,
|
|
"mean_token_accuracy": 0.19703703224658967,
|
|
"num_tokens": 8961611.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 4.937661504745483,
|
|
"epoch": 3.7069988137603795,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041498708302375277,
|
|
"loss": 4.7334,
|
|
"mean_token_accuracy": 0.20828222185373307,
|
|
"num_tokens": 8975993.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.019860696792603,
|
|
"epoch": 3.7129300118623965,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004146138655031165,
|
|
"loss": 4.8078,
|
|
"mean_token_accuracy": 0.1950700029730797,
|
|
"num_tokens": 8989198.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.036433696746826,
|
|
"epoch": 3.718861209964413,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004142400222908681,
|
|
"loss": 4.8078,
|
|
"mean_token_accuracy": 0.20096368491649627,
|
|
"num_tokens": 9004592.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 4.998563623428344,
|
|
"epoch": 3.7247924080664294,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004138655550624183,
|
|
"loss": 4.773,
|
|
"mean_token_accuracy": 0.19526535868644715,
|
|
"num_tokens": 9017772.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.036626148223877,
|
|
"epoch": 3.730723606168446,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004134904654959745,
|
|
"loss": 4.7073,
|
|
"mean_token_accuracy": 0.20760386288166047,
|
|
"num_tokens": 9031182.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.066286182403564,
|
|
"epoch": 3.7366548042704624,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041311475527253284,
|
|
"loss": 4.7977,
|
|
"mean_token_accuracy": 0.19728391021490096,
|
|
"num_tokens": 9044934.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.016174173355102,
|
|
"epoch": 3.7425860023724793,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041273842607587137,
|
|
"loss": 4.7928,
|
|
"mean_token_accuracy": 0.1950550675392151,
|
|
"num_tokens": 9059795.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 4.972341299057007,
|
|
"epoch": 3.7485172004744958,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041236147959254166,
|
|
"loss": 4.782,
|
|
"mean_token_accuracy": 0.20030613243579865,
|
|
"num_tokens": 9073410.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.022267293930054,
|
|
"epoch": 3.7544483985765122,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041198391751186227,
|
|
"loss": 4.8398,
|
|
"mean_token_accuracy": 0.19292792528867722,
|
|
"num_tokens": 9089338.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.042642307281494,
|
|
"epoch": 3.760379596678529,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004116057415259101,
|
|
"loss": 4.7528,
|
|
"mean_token_accuracy": 0.20497467815876008,
|
|
"num_tokens": 9103520.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.021176958084107,
|
|
"epoch": 3.7663107947805456,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041122695332951363,
|
|
"loss": 4.7706,
|
|
"mean_token_accuracy": 0.20218247920274734,
|
|
"num_tokens": 9117935.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 4.987675666809082,
|
|
"epoch": 3.772241992882562,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004108475546202449,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.20431260019540787,
|
|
"num_tokens": 9132457.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.050307130813598,
|
|
"epoch": 3.778173190984579,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004104675470984121,
|
|
"loss": 4.8232,
|
|
"mean_token_accuracy": 0.19753638207912444,
|
|
"num_tokens": 9147329.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.029834890365601,
|
|
"epoch": 3.7841043890865955,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004100869324670517,
|
|
"loss": 4.7634,
|
|
"mean_token_accuracy": 0.20562210381031037,
|
|
"num_tokens": 9161052.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.00586576461792,
|
|
"epoch": 3.790035587188612,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004097057124319212,
|
|
"loss": 4.8242,
|
|
"mean_token_accuracy": 0.20105937719345093,
|
|
"num_tokens": 9175900.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 4.9953101634979244,
|
|
"epoch": 3.795966785290629,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000409323888701491,
|
|
"loss": 4.7846,
|
|
"mean_token_accuracy": 0.2029906615614891,
|
|
"num_tokens": 9189706.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.035350704193116,
|
|
"epoch": 3.8018979833926454,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004089414629869373,
|
|
"loss": 4.8483,
|
|
"mean_token_accuracy": 0.1962611809372902,
|
|
"num_tokens": 9203727.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.05778546333313,
|
|
"epoch": 3.807829181494662,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00040855843700213386,
|
|
"loss": 4.796,
|
|
"mean_token_accuracy": 0.19637212604284288,
|
|
"num_tokens": 9217664.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 4.937251186370849,
|
|
"epoch": 3.813760379596679,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004081748124636449,
|
|
"loss": 4.7115,
|
|
"mean_token_accuracy": 0.20479025095701217,
|
|
"num_tokens": 9233474.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 4.995423221588135,
|
|
"epoch": 3.8196915776986953,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040779059109071706,
|
|
"loss": 4.7895,
|
|
"mean_token_accuracy": 0.20069828033447265,
|
|
"num_tokens": 9247567.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.027128553390503,
|
|
"epoch": 3.8256227758007118,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004074057746052713,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.1999478057026863,
|
|
"num_tokens": 9262264.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.114210844039917,
|
|
"epoch": 3.8315539739027282,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004070203647318963,
|
|
"loss": 4.9095,
|
|
"mean_token_accuracy": 0.18577013909816742,
|
|
"num_tokens": 9276893.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.053824806213379,
|
|
"epoch": 3.8374851720047447,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040663436319783965,
|
|
"loss": 4.8054,
|
|
"mean_token_accuracy": 0.19727878272533417,
|
|
"num_tokens": 9291646.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.021931886672974,
|
|
"epoch": 3.8434163701067616,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004062477717330006,
|
|
"loss": 4.8344,
|
|
"mean_token_accuracy": 0.19990530163049697,
|
|
"num_tokens": 9305789.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 4.9833355903625485,
|
|
"epoch": 3.849347568208778,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040586059206992214,
|
|
"loss": 4.7497,
|
|
"mean_token_accuracy": 0.1978788733482361,
|
|
"num_tokens": 9321039.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.031898260116577,
|
|
"epoch": 3.8552787663107946,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004054728259437836,
|
|
"loss": 4.82,
|
|
"mean_token_accuracy": 0.1935506910085678,
|
|
"num_tokens": 9337105.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.07962851524353,
|
|
"epoch": 3.8612099644128115,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040508447509239224,
|
|
"loss": 4.8061,
|
|
"mean_token_accuracy": 0.19505731612443925,
|
|
"num_tokens": 9350684.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.032706165313721,
|
|
"epoch": 3.867141162514828,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000404695541256176,
|
|
"loss": 4.8298,
|
|
"mean_token_accuracy": 0.19414604753255843,
|
|
"num_tokens": 9365074.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.051387643814087,
|
|
"epoch": 3.8730723606168445,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004043060261781756,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.18420381844043732,
|
|
"num_tokens": 9380912.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.086946487426758,
|
|
"epoch": 3.8790035587188614,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040391593160403656,
|
|
"loss": 4.8248,
|
|
"mean_token_accuracy": 0.20802772492170335,
|
|
"num_tokens": 9393812.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 4.938499832153321,
|
|
"epoch": 3.884934756820878,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040352525928200137,
|
|
"loss": 4.7925,
|
|
"mean_token_accuracy": 0.20416243374347687,
|
|
"num_tokens": 9409731.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.003332328796387,
|
|
"epoch": 3.8908659549228943,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004031340109629017,
|
|
"loss": 4.7663,
|
|
"mean_token_accuracy": 0.20782396644353868,
|
|
"num_tokens": 9422743.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.017239379882812,
|
|
"epoch": 3.8967971530249113,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000402742188400151,
|
|
"loss": 4.8235,
|
|
"mean_token_accuracy": 0.19541184455156327,
|
|
"num_tokens": 9436440.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 4.970128440856934,
|
|
"epoch": 3.9027283511269277,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040234979334973577,
|
|
"loss": 4.8555,
|
|
"mean_token_accuracy": 0.19211765229701996,
|
|
"num_tokens": 9451668.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.056919956207276,
|
|
"epoch": 3.908659549228944,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040195682757020846,
|
|
"loss": 4.7723,
|
|
"mean_token_accuracy": 0.21034520268440246,
|
|
"num_tokens": 9464700.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 4.960184717178345,
|
|
"epoch": 3.914590747330961,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040156329282267915,
|
|
"loss": 4.7365,
|
|
"mean_token_accuracy": 0.20848041474819184,
|
|
"num_tokens": 9479003.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 4.990036296844482,
|
|
"epoch": 3.9205219454329776,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00040116919087080787,
|
|
"loss": 4.7514,
|
|
"mean_token_accuracy": 0.20134748965501786,
|
|
"num_tokens": 9493134.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.011323022842407,
|
|
"epoch": 3.926453143534994,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040077452348079664,
|
|
"loss": 4.7617,
|
|
"mean_token_accuracy": 0.20163092166185378,
|
|
"num_tokens": 9507087.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.019417953491211,
|
|
"epoch": 3.9323843416370106,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004003792924213814,
|
|
"loss": 4.7621,
|
|
"mean_token_accuracy": 0.20881457775831222,
|
|
"num_tokens": 9518328.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 4.98180365562439,
|
|
"epoch": 3.938315539739027,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003999834994638243,
|
|
"loss": 4.8216,
|
|
"mean_token_accuracy": 0.20216355621814727,
|
|
"num_tokens": 9532413.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.001606082916259,
|
|
"epoch": 3.944246737841044,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003995871463819059,
|
|
"loss": 4.761,
|
|
"mean_token_accuracy": 0.2026512235403061,
|
|
"num_tokens": 9547681.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 4.967384576797485,
|
|
"epoch": 3.9501779359430604,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039919023495191665,
|
|
"loss": 4.7276,
|
|
"mean_token_accuracy": 0.20670733749866485,
|
|
"num_tokens": 9560198.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 4.987757778167724,
|
|
"epoch": 3.956109134045077,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003987927669526494,
|
|
"loss": 4.8062,
|
|
"mean_token_accuracy": 0.1983866959810257,
|
|
"num_tokens": 9574347.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 4.972973728179932,
|
|
"epoch": 3.962040332147094,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039839474416539163,
|
|
"loss": 4.7004,
|
|
"mean_token_accuracy": 0.20886301398277282,
|
|
"num_tokens": 9587337.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 4.984016704559326,
|
|
"epoch": 3.9679715302491103,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039799616837391655,
|
|
"loss": 4.784,
|
|
"mean_token_accuracy": 0.19981919676065446,
|
|
"num_tokens": 9601047.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.0524254322052,
|
|
"epoch": 3.973902728351127,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039759704136447627,
|
|
"loss": 4.7662,
|
|
"mean_token_accuracy": 0.20110076069831848,
|
|
"num_tokens": 9614164.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.00496735572815,
|
|
"epoch": 3.9798339264531437,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003971973649257929,
|
|
"loss": 4.8026,
|
|
"mean_token_accuracy": 0.19617463350296022,
|
|
"num_tokens": 9628179.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 4.986306190490723,
|
|
"epoch": 3.98576512455516,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000396797140849051,
|
|
"loss": 4.8062,
|
|
"mean_token_accuracy": 0.20138847976922988,
|
|
"num_tokens": 9641575.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.002590847015381,
|
|
"epoch": 3.9916963226571767,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039639637092788935,
|
|
"loss": 4.7214,
|
|
"mean_token_accuracy": 0.20492513924837114,
|
|
"num_tokens": 9655381.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.0504356861114506,
|
|
"epoch": 3.9976275207591936,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000395995056958393,
|
|
"loss": 4.8149,
|
|
"mean_token_accuracy": 0.19649633914232253,
|
|
"num_tokens": 9670477.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 4.922432947158813,
|
|
"epoch": 4.00355871886121,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039559320073908526,
|
|
"loss": 4.5875,
|
|
"mean_token_accuracy": 0.2211996853351593,
|
|
"num_tokens": 9683964.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.0110671043396,
|
|
"epoch": 4.0094899169632265,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00039519080407091956,
|
|
"loss": 4.58,
|
|
"mean_token_accuracy": 0.21853467971086502,
|
|
"num_tokens": 9698415.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.0225770473480225,
|
|
"epoch": 4.0154211150652435,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039478786875727116,
|
|
"loss": 4.5708,
|
|
"mean_token_accuracy": 0.21765107959508895,
|
|
"num_tokens": 9712895.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 4.9370323657989506,
|
|
"epoch": 4.0213523131672595,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039438439660392983,
|
|
"loss": 4.475,
|
|
"mean_token_accuracy": 0.22428351193666457,
|
|
"num_tokens": 9728872.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 4.997824096679688,
|
|
"epoch": 4.027283511269276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039398038941909066,
|
|
"loss": 4.5467,
|
|
"mean_token_accuracy": 0.21755390614271164,
|
|
"num_tokens": 9744284.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 4.97338900566101,
|
|
"epoch": 4.033214709371293,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000393575849013347,
|
|
"loss": 4.5463,
|
|
"mean_token_accuracy": 0.2152925118803978,
|
|
"num_tokens": 9759374.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 4.930527114868164,
|
|
"epoch": 4.039145907473309,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039317077719968143,
|
|
"loss": 4.5183,
|
|
"mean_token_accuracy": 0.21953357756137848,
|
|
"num_tokens": 9773919.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 4.989052486419678,
|
|
"epoch": 4.045077105575326,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039276517579345846,
|
|
"loss": 4.5668,
|
|
"mean_token_accuracy": 0.21412246227264403,
|
|
"num_tokens": 9788023.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.021774053573608,
|
|
"epoch": 4.051008303677343,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003923590466124159,
|
|
"loss": 4.561,
|
|
"mean_token_accuracy": 0.21579589694738388,
|
|
"num_tokens": 9803037.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 4.994525671005249,
|
|
"epoch": 4.056939501779359,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039195239147665664,
|
|
"loss": 4.5317,
|
|
"mean_token_accuracy": 0.21720170676708223,
|
|
"num_tokens": 9816782.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 4.995402812957764,
|
|
"epoch": 4.062870699881376,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039154521220864094,
|
|
"loss": 4.544,
|
|
"mean_token_accuracy": 0.21858830749988556,
|
|
"num_tokens": 9830471.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 4.991329097747803,
|
|
"epoch": 4.068801897983393,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039113751063317795,
|
|
"loss": 4.597,
|
|
"mean_token_accuracy": 0.21205688118934632,
|
|
"num_tokens": 9844303.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 4.940326738357544,
|
|
"epoch": 4.074733096085409,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003907292885774175,
|
|
"loss": 4.5388,
|
|
"mean_token_accuracy": 0.21253766417503356,
|
|
"num_tokens": 9861002.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 4.897701930999756,
|
|
"epoch": 4.080664294187426,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003903205478708419,
|
|
"loss": 4.5072,
|
|
"mean_token_accuracy": 0.22263137847185135,
|
|
"num_tokens": 9876338.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.052014923095703,
|
|
"epoch": 4.086595492289442,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003899112903452582,
|
|
"loss": 4.6115,
|
|
"mean_token_accuracy": 0.21495201140642167,
|
|
"num_tokens": 9891217.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 4.955441427230835,
|
|
"epoch": 4.092526690391459,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950151783478926,
|
|
"loss": 4.547,
|
|
"mean_token_accuracy": 0.22495205700397491,
|
|
"num_tokens": 9905309.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 4.965147495269775,
|
|
"epoch": 4.098457888493476,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000389091232175866,
|
|
"loss": 4.5015,
|
|
"mean_token_accuracy": 0.22329854369163513,
|
|
"num_tokens": 9918347.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 4.948949956893921,
|
|
"epoch": 4.104389086595492,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003886804352072192,
|
|
"loss": 4.5585,
|
|
"mean_token_accuracy": 0.21841878592967987,
|
|
"num_tokens": 9933068.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 4.864326620101929,
|
|
"epoch": 4.110320284697509,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003882691287698709,
|
|
"loss": 4.5283,
|
|
"mean_token_accuracy": 0.2233476459980011,
|
|
"num_tokens": 9948149.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 4.933009481430053,
|
|
"epoch": 4.116251482799526,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003878573147071265,
|
|
"loss": 4.5161,
|
|
"mean_token_accuracy": 0.2156417414546013,
|
|
"num_tokens": 9961494.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 4.932680368423462,
|
|
"epoch": 4.122182680901542,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003874449948645664,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.2222415953874588,
|
|
"num_tokens": 9977270.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 4.86724328994751,
|
|
"epoch": 4.128113879003559,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003870321710900375,
|
|
"loss": 4.4522,
|
|
"mean_token_accuracy": 0.2294168159365654,
|
|
"num_tokens": 9991102.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 4.947033166885376,
|
|
"epoch": 4.134045077105576,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038661884523364537,
|
|
"loss": 4.5741,
|
|
"mean_token_accuracy": 0.22195196747779847,
|
|
"num_tokens": 10004991.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 4.9410834312438965,
|
|
"epoch": 4.139976275207592,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038620501914774554,
|
|
"loss": 4.5944,
|
|
"mean_token_accuracy": 0.2187734752893448,
|
|
"num_tokens": 10019654.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 4.893217134475708,
|
|
"epoch": 4.145907473309609,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038579069468693526,
|
|
"loss": 4.4958,
|
|
"mean_token_accuracy": 0.22282861918210983,
|
|
"num_tokens": 10033831.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 4.945186996459961,
|
|
"epoch": 4.1518386714116255,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00038537587370804567,
|
|
"loss": 4.5099,
|
|
"mean_token_accuracy": 0.21789855360984803,
|
|
"num_tokens": 10049549.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 4.1518386714116255,
|
|
"eval_entropy": 4.839259713705332,
|
|
"eval_loss": 5.367321491241455,
|
|
"eval_mean_token_accuracy": 0.18195566196748816,
|
|
"eval_num_tokens": 10049549.0,
|
|
"eval_runtime": 2.4857,
|
|
"eval_samples_per_second": 1049.215,
|
|
"eval_steps_per_second": 131.152,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 8420,
|
|
"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": 1.771285416192e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|