2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.1862396204033214,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1000,
|
||
|
|
"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
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"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": 5083861082112000.0,
|
||
|
|
"train_batch_size": 16,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|