6035 lines
164 KiB
JSON
6035 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.06703386327326355,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.7426682472229,
|
|
"epoch": 0.00011172310545543924,
|
|
"grad_norm": 6.25,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.5231,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 4250.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742681884765625,
|
|
"epoch": 0.00022344621091087847,
|
|
"grad_norm": 7.03125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.4891,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8228.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742702770233155,
|
|
"epoch": 0.0003351693163663177,
|
|
"grad_norm": 6.8125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4445,
|
|
"mean_token_accuracy": 0.0003105590119957924,
|
|
"num_tokens": 12209.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742706871032714,
|
|
"epoch": 0.00044689242182175694,
|
|
"grad_norm": 6.15625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3987,
|
|
"mean_token_accuracy": 0.0004866180010139942,
|
|
"num_tokens": 16225.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742706108093262,
|
|
"epoch": 0.0005586155272771962,
|
|
"grad_norm": 5.53125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.2787,
|
|
"mean_token_accuracy": 0.010022059944458307,
|
|
"num_tokens": 20212.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742425346374512,
|
|
"epoch": 0.0006703386327326354,
|
|
"grad_norm": 5.0,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.173,
|
|
"mean_token_accuracy": 0.04117634426802397,
|
|
"num_tokens": 24445.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741415786743165,
|
|
"epoch": 0.0007820617381880747,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9901,
|
|
"mean_token_accuracy": 0.052767305821180346,
|
|
"num_tokens": 28365.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.739409732818604,
|
|
"epoch": 0.0008937848436435139,
|
|
"grad_norm": 3.578125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.9621,
|
|
"mean_token_accuracy": 0.052475782483816145,
|
|
"num_tokens": 33055.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.7363787651062,
|
|
"epoch": 0.0010055079490989532,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.811,
|
|
"mean_token_accuracy": 0.062037082761526106,
|
|
"num_tokens": 37599.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.733420372009277,
|
|
"epoch": 0.0011172310545543925,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6744,
|
|
"mean_token_accuracy": 0.06838746592402459,
|
|
"num_tokens": 41934.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.731625938415528,
|
|
"epoch": 0.0012289541600098315,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6365,
|
|
"mean_token_accuracy": 0.05915887728333473,
|
|
"num_tokens": 46178.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.729843425750733,
|
|
"epoch": 0.0013406772654652708,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5494,
|
|
"mean_token_accuracy": 0.0692246112972498,
|
|
"num_tokens": 50513.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.727421474456786,
|
|
"epoch": 0.0014524003709207101,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5028,
|
|
"mean_token_accuracy": 0.0702465757727623,
|
|
"num_tokens": 54924.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.722854518890381,
|
|
"epoch": 0.0015641234763761494,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4107,
|
|
"mean_token_accuracy": 0.06344567574560642,
|
|
"num_tokens": 59083.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.71631669998169,
|
|
"epoch": 0.0016758465818315885,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3233,
|
|
"mean_token_accuracy": 0.06774163469672204,
|
|
"num_tokens": 63324.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.705671787261963,
|
|
"epoch": 0.0017875696872870278,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.3567,
|
|
"mean_token_accuracy": 0.06332114227116108,
|
|
"num_tokens": 67738.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.693737983703613,
|
|
"epoch": 0.001899292792742467,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.2866,
|
|
"mean_token_accuracy": 0.06349676214158535,
|
|
"num_tokens": 72305.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.675388622283936,
|
|
"epoch": 0.0020110158981979064,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0821,
|
|
"mean_token_accuracy": 0.06834135130047798,
|
|
"num_tokens": 76579.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.65205717086792,
|
|
"epoch": 0.0021227390036533456,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.0421,
|
|
"mean_token_accuracy": 0.06946654319763183,
|
|
"num_tokens": 80812.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.615971279144286,
|
|
"epoch": 0.002234462109108785,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.9523,
|
|
"mean_token_accuracy": 0.06732719540596008,
|
|
"num_tokens": 85090.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.59008207321167,
|
|
"epoch": 0.0023461852145642242,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.888,
|
|
"mean_token_accuracy": 0.06908667460083961,
|
|
"num_tokens": 89578.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.536420917510986,
|
|
"epoch": 0.002457908320019663,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.789,
|
|
"mean_token_accuracy": 0.0728236336261034,
|
|
"num_tokens": 94117.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.488511657714843,
|
|
"epoch": 0.0025696314254751024,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6132,
|
|
"mean_token_accuracy": 0.07127482630312443,
|
|
"num_tokens": 98082.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.439968013763428,
|
|
"epoch": 0.0026813545309305417,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.5714,
|
|
"mean_token_accuracy": 0.07672090865671635,
|
|
"num_tokens": 102327.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.355792045593262,
|
|
"epoch": 0.002793077636385981,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.4426,
|
|
"mean_token_accuracy": 0.0740627009421587,
|
|
"num_tokens": 106567.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.286309623718262,
|
|
"epoch": 0.0029048007418414202,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.3003,
|
|
"mean_token_accuracy": 0.07362989187240601,
|
|
"num_tokens": 110654.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.204053020477295,
|
|
"epoch": 0.0030165238472968595,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.2511,
|
|
"mean_token_accuracy": 0.062000279501080516,
|
|
"num_tokens": 114679.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.102067852020264,
|
|
"epoch": 0.003128246952752299,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.1849,
|
|
"mean_token_accuracy": 0.06811538599431514,
|
|
"num_tokens": 118817.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.926943397521972,
|
|
"epoch": 0.003239970058207738,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.0767,
|
|
"mean_token_accuracy": 0.06979594528675079,
|
|
"num_tokens": 123188.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.793034744262695,
|
|
"epoch": 0.003351693163663177,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.981,
|
|
"mean_token_accuracy": 0.06847230046987533,
|
|
"num_tokens": 127767.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.643688774108886,
|
|
"epoch": 0.0034634162691186163,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7945,
|
|
"mean_token_accuracy": 0.06945906654000282,
|
|
"num_tokens": 131837.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.430543518066406,
|
|
"epoch": 0.0035751393745740555,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7734,
|
|
"mean_token_accuracy": 0.07027286775410176,
|
|
"num_tokens": 136247.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.239261722564697,
|
|
"epoch": 0.003686862480029495,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5788,
|
|
"mean_token_accuracy": 0.07950169630348683,
|
|
"num_tokens": 140170.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.976140880584717,
|
|
"epoch": 0.003798585585484934,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6177,
|
|
"mean_token_accuracy": 0.07785017378628253,
|
|
"num_tokens": 144139.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.843453693389893,
|
|
"epoch": 0.003910308690940373,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.5659,
|
|
"mean_token_accuracy": 0.07487303391098976,
|
|
"num_tokens": 148792.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.658325004577637,
|
|
"epoch": 0.004022031796395813,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4988,
|
|
"mean_token_accuracy": 0.07942216768860817,
|
|
"num_tokens": 152844.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.59526195526123,
|
|
"epoch": 0.0041337549018512516,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.527,
|
|
"mean_token_accuracy": 0.07417443916201591,
|
|
"num_tokens": 157366.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.467089462280274,
|
|
"epoch": 0.004245478007306691,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3623,
|
|
"mean_token_accuracy": 0.07755868881940842,
|
|
"num_tokens": 161348.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.307873630523682,
|
|
"epoch": 0.00435720111276213,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3815,
|
|
"mean_token_accuracy": 0.08716461397707462,
|
|
"num_tokens": 165647.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.236515140533447,
|
|
"epoch": 0.00446892421821757,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2754,
|
|
"mean_token_accuracy": 0.08076057620346547,
|
|
"num_tokens": 169521.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.256762790679932,
|
|
"epoch": 0.004580647323673009,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3426,
|
|
"mean_token_accuracy": 0.0812241055071354,
|
|
"num_tokens": 173466.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.131280899047852,
|
|
"epoch": 0.0046923704291284484,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2826,
|
|
"mean_token_accuracy": 0.07643571458756923,
|
|
"num_tokens": 177663.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.097990989685059,
|
|
"epoch": 0.004804093534583887,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2745,
|
|
"mean_token_accuracy": 0.08235705867409707,
|
|
"num_tokens": 181778.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.089111948013306,
|
|
"epoch": 0.004915816640039326,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.2736,
|
|
"mean_token_accuracy": 0.08633389472961425,
|
|
"num_tokens": 185525.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.083420944213866,
|
|
"epoch": 0.005027539745494766,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.153,
|
|
"mean_token_accuracy": 0.08806331530213356,
|
|
"num_tokens": 189418.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.933328151702881,
|
|
"epoch": 0.005139262850950205,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2217,
|
|
"mean_token_accuracy": 0.08842612579464912,
|
|
"num_tokens": 193494.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.018900680541993,
|
|
"epoch": 0.0052509859564056445,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2661,
|
|
"mean_token_accuracy": 0.08137304298579692,
|
|
"num_tokens": 198018.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.955441856384278,
|
|
"epoch": 0.005362709061861083,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1847,
|
|
"mean_token_accuracy": 0.08625513166189194,
|
|
"num_tokens": 202296.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.9594367980957035,
|
|
"epoch": 0.005474432167316523,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1706,
|
|
"mean_token_accuracy": 0.08195730969309807,
|
|
"num_tokens": 206694.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.792031574249267,
|
|
"epoch": 0.005586155272771962,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2007,
|
|
"mean_token_accuracy": 0.08904931843280792,
|
|
"num_tokens": 210810.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.920461797714234,
|
|
"epoch": 0.005697878378227402,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1818,
|
|
"mean_token_accuracy": 0.0905133418738842,
|
|
"num_tokens": 215044.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.8493430614471436,
|
|
"epoch": 0.0058096014836828405,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.28,
|
|
"mean_token_accuracy": 0.08591654896736145,
|
|
"num_tokens": 219235.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.84934287071228,
|
|
"epoch": 0.005921324589138279,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.0922,
|
|
"mean_token_accuracy": 0.0903876356780529,
|
|
"num_tokens": 223639.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.785561227798462,
|
|
"epoch": 0.006033047694593719,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1258,
|
|
"mean_token_accuracy": 0.09057728350162506,
|
|
"num_tokens": 227873.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.707937574386596,
|
|
"epoch": 0.006144770800049158,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0661,
|
|
"mean_token_accuracy": 0.09807337448000908,
|
|
"num_tokens": 232147.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.739069509506225,
|
|
"epoch": 0.006256493905504598,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1358,
|
|
"mean_token_accuracy": 0.09250000454485416,
|
|
"num_tokens": 236456.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.7190714359283445,
|
|
"epoch": 0.0063682170109600365,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1583,
|
|
"mean_token_accuracy": 0.09051149562001229,
|
|
"num_tokens": 241039.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.938947439193726,
|
|
"epoch": 0.006479940116415476,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1915,
|
|
"mean_token_accuracy": 0.08653632178902626,
|
|
"num_tokens": 245132.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.673107481002807,
|
|
"epoch": 0.006591663221870915,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0872,
|
|
"mean_token_accuracy": 0.09988043382763863,
|
|
"num_tokens": 249152.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.712965631484986,
|
|
"epoch": 0.006703386327326354,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0503,
|
|
"mean_token_accuracy": 0.09596830010414123,
|
|
"num_tokens": 253439.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.6600532054901125,
|
|
"epoch": 0.006815109432781794,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0731,
|
|
"mean_token_accuracy": 0.09302671104669571,
|
|
"num_tokens": 258066.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.665358448028565,
|
|
"epoch": 0.0069268325382372325,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0332,
|
|
"mean_token_accuracy": 0.0973147690296173,
|
|
"num_tokens": 261954.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.616210794448852,
|
|
"epoch": 0.007038555643692672,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0779,
|
|
"mean_token_accuracy": 0.10462095588445663,
|
|
"num_tokens": 266650.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.689846324920654,
|
|
"epoch": 0.007150278749148111,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1433,
|
|
"mean_token_accuracy": 0.09891897812485695,
|
|
"num_tokens": 271069.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.705677938461304,
|
|
"epoch": 0.007262001854603551,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0039,
|
|
"mean_token_accuracy": 0.10242248028516769,
|
|
"num_tokens": 275084.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.603102445602417,
|
|
"epoch": 0.00737372496005899,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0745,
|
|
"mean_token_accuracy": 0.1031483568251133,
|
|
"num_tokens": 279721.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.619607782363891,
|
|
"epoch": 0.007485448065514429,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0708,
|
|
"mean_token_accuracy": 0.10527726709842682,
|
|
"num_tokens": 284317.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.600710487365722,
|
|
"epoch": 0.007597171170969868,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0451,
|
|
"mean_token_accuracy": 0.10766607597470283,
|
|
"num_tokens": 288870.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.61973729133606,
|
|
"epoch": 0.007708894276425307,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9812,
|
|
"mean_token_accuracy": 0.11351362988352776,
|
|
"num_tokens": 292996.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.5854551792144775,
|
|
"epoch": 0.007820617381880746,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9806,
|
|
"mean_token_accuracy": 0.10384939089417458,
|
|
"num_tokens": 297238.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.531381893157959,
|
|
"epoch": 0.007932340487336187,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9793,
|
|
"mean_token_accuracy": 0.1117280475795269,
|
|
"num_tokens": 301453.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.653309726715088,
|
|
"epoch": 0.008044063592791625,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9327,
|
|
"mean_token_accuracy": 0.10786554217338562,
|
|
"num_tokens": 305949.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.534815788269043,
|
|
"epoch": 0.008155786698247064,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9583,
|
|
"mean_token_accuracy": 0.11513907313346863,
|
|
"num_tokens": 310097.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.547474193572998,
|
|
"epoch": 0.008267509803702503,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9517,
|
|
"mean_token_accuracy": 0.10539396926760673,
|
|
"num_tokens": 314567.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.457708692550659,
|
|
"epoch": 0.008379232909157944,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0323,
|
|
"mean_token_accuracy": 0.10927818715572357,
|
|
"num_tokens": 319166.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.515052604675293,
|
|
"epoch": 0.008490956014613383,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9204,
|
|
"mean_token_accuracy": 0.11223233640193939,
|
|
"num_tokens": 323682.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.488761281967163,
|
|
"epoch": 0.008602679120068821,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8299,
|
|
"mean_token_accuracy": 0.12143486216664315,
|
|
"num_tokens": 327994.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.412152099609375,
|
|
"epoch": 0.00871440222552426,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9058,
|
|
"mean_token_accuracy": 0.11854805946350097,
|
|
"num_tokens": 332026.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.578387832641601,
|
|
"epoch": 0.0088261253309797,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9475,
|
|
"mean_token_accuracy": 0.11811894848942757,
|
|
"num_tokens": 336552.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.504688882827759,
|
|
"epoch": 0.00893784843643514,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8884,
|
|
"mean_token_accuracy": 0.1116393692791462,
|
|
"num_tokens": 340643.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.4580738067626955,
|
|
"epoch": 0.009049571541890579,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8029,
|
|
"mean_token_accuracy": 0.12075437754392623,
|
|
"num_tokens": 344886.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.3586314678192135,
|
|
"epoch": 0.009161294647346017,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8632,
|
|
"mean_token_accuracy": 0.1191755935549736,
|
|
"num_tokens": 349115.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.527571535110473,
|
|
"epoch": 0.009273017752801456,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8235,
|
|
"mean_token_accuracy": 0.12523479163646697,
|
|
"num_tokens": 353368.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.509571599960327,
|
|
"epoch": 0.009384740858256897,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8657,
|
|
"mean_token_accuracy": 0.10757644474506378,
|
|
"num_tokens": 357382.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.441834354400635,
|
|
"epoch": 0.009496463963712336,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7799,
|
|
"mean_token_accuracy": 0.12460733354091644,
|
|
"num_tokens": 361542.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.419776153564453,
|
|
"epoch": 0.009608187069167775,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.84,
|
|
"mean_token_accuracy": 0.11263928636908531,
|
|
"num_tokens": 366006.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.367758464813233,
|
|
"epoch": 0.009719910174623213,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7506,
|
|
"mean_token_accuracy": 0.12978531718254088,
|
|
"num_tokens": 370021.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.408233594894409,
|
|
"epoch": 0.009831633280078652,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8773,
|
|
"mean_token_accuracy": 0.12414649501442909,
|
|
"num_tokens": 374434.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.441655158996582,
|
|
"epoch": 0.009943356385534093,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9027,
|
|
"mean_token_accuracy": 0.11324851140379906,
|
|
"num_tokens": 378634.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.315918588638306,
|
|
"epoch": 0.010055079490989532,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7869,
|
|
"mean_token_accuracy": 0.1252473659813404,
|
|
"num_tokens": 382904.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.470485210418701,
|
|
"epoch": 0.01016680259644497,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7635,
|
|
"mean_token_accuracy": 0.12090180814266205,
|
|
"num_tokens": 386970.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.348088216781616,
|
|
"epoch": 0.01027852570190041,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.729,
|
|
"mean_token_accuracy": 0.13390202075242996,
|
|
"num_tokens": 391043.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.392842721939087,
|
|
"epoch": 0.01039024880735585,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.13383440747857095,
|
|
"num_tokens": 395413.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.40152382850647,
|
|
"epoch": 0.010501971912811289,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8385,
|
|
"mean_token_accuracy": 0.12566340565681458,
|
|
"num_tokens": 399821.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.2655271053314205,
|
|
"epoch": 0.010613695018266728,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6582,
|
|
"mean_token_accuracy": 0.13715523406863211,
|
|
"num_tokens": 404043.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.422811889648438,
|
|
"epoch": 0.010725418123722167,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.803,
|
|
"mean_token_accuracy": 0.1260749615728855,
|
|
"num_tokens": 408339.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.263138484954834,
|
|
"epoch": 0.010837141229177606,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6856,
|
|
"mean_token_accuracy": 0.13459724336862564,
|
|
"num_tokens": 412384.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.3362548828125,
|
|
"epoch": 0.010948864334633046,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.8108,
|
|
"mean_token_accuracy": 0.12614913210272788,
|
|
"num_tokens": 416891.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.36023063659668,
|
|
"epoch": 0.011060587440088485,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6642,
|
|
"mean_token_accuracy": 0.1329216368496418,
|
|
"num_tokens": 420980.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.2991362571716305,
|
|
"epoch": 0.011172310545543924,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.7212,
|
|
"mean_token_accuracy": 0.13121648952364923,
|
|
"num_tokens": 425454.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.339645147323608,
|
|
"epoch": 0.011284033650999363,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7456,
|
|
"mean_token_accuracy": 0.12466516643762589,
|
|
"num_tokens": 430087.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.289543581008911,
|
|
"epoch": 0.011395756756454803,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.674,
|
|
"mean_token_accuracy": 0.13790097907185556,
|
|
"num_tokens": 434062.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.032429838180542,
|
|
"epoch": 0.011507479861910242,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6567,
|
|
"mean_token_accuracy": 0.13795162215828896,
|
|
"num_tokens": 437882.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.306787776947021,
|
|
"epoch": 0.011619202967365681,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6863,
|
|
"mean_token_accuracy": 0.13145707920193672,
|
|
"num_tokens": 442248.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.304118633270264,
|
|
"epoch": 0.01173092607282112,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.6543,
|
|
"mean_token_accuracy": 0.13509279638528823,
|
|
"num_tokens": 446492.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.187564706802368,
|
|
"epoch": 0.011842649178276559,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6878,
|
|
"mean_token_accuracy": 0.1363551899790764,
|
|
"num_tokens": 450359.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.3090503215789795,
|
|
"epoch": 0.011954372283732,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6973,
|
|
"mean_token_accuracy": 0.13162412643432617,
|
|
"num_tokens": 454508.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.349436283111572,
|
|
"epoch": 0.012066095389187438,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.666,
|
|
"mean_token_accuracy": 0.133739610016346,
|
|
"num_tokens": 459028.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.165412759780883,
|
|
"epoch": 0.012177818494642877,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6084,
|
|
"mean_token_accuracy": 0.13254478350281715,
|
|
"num_tokens": 462911.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.141500329971313,
|
|
"epoch": 0.012289541600098316,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6564,
|
|
"mean_token_accuracy": 0.13796778842806817,
|
|
"num_tokens": 467129.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.303054237365723,
|
|
"epoch": 0.012401264705553756,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5808,
|
|
"mean_token_accuracy": 0.139659858494997,
|
|
"num_tokens": 471324.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.217333889007568,
|
|
"epoch": 0.012512987811009195,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7315,
|
|
"mean_token_accuracy": 0.12687695473432542,
|
|
"num_tokens": 475761.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.198446226119995,
|
|
"epoch": 0.012624710916464634,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5304,
|
|
"mean_token_accuracy": 0.13995881900191307,
|
|
"num_tokens": 479532.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.236463642120361,
|
|
"epoch": 0.012736434021920073,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6868,
|
|
"mean_token_accuracy": 0.13790254518389702,
|
|
"num_tokens": 483844.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.325732755661011,
|
|
"epoch": 0.012848157127375512,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7171,
|
|
"mean_token_accuracy": 0.1334671013057232,
|
|
"num_tokens": 488402.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.063196992874145,
|
|
"epoch": 0.012959880232830952,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4612,
|
|
"mean_token_accuracy": 0.14302808195352554,
|
|
"num_tokens": 492473.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.110111331939697,
|
|
"epoch": 0.013071603338286391,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.568,
|
|
"mean_token_accuracy": 0.14078054577112198,
|
|
"num_tokens": 496830.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.2420226573944095,
|
|
"epoch": 0.01318332644374183,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.631,
|
|
"mean_token_accuracy": 0.13996972143650055,
|
|
"num_tokens": 500996.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.113031435012817,
|
|
"epoch": 0.013295049549197269,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5937,
|
|
"mean_token_accuracy": 0.13735369965434074,
|
|
"num_tokens": 505174.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.149052238464355,
|
|
"epoch": 0.013406772654652708,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5965,
|
|
"mean_token_accuracy": 0.14343740195035934,
|
|
"num_tokens": 509425.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.0243888854980465,
|
|
"epoch": 0.013518495760108148,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4748,
|
|
"mean_token_accuracy": 0.14360842779278754,
|
|
"num_tokens": 513271.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.198608922958374,
|
|
"epoch": 0.013630218865563587,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.641,
|
|
"mean_token_accuracy": 0.146073829382658,
|
|
"num_tokens": 517879.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.045837879180908,
|
|
"epoch": 0.013741941971019026,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.13332991302013397,
|
|
"num_tokens": 522321.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.1533918380737305,
|
|
"epoch": 0.013853665076474465,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4453,
|
|
"mean_token_accuracy": 0.15030871629714965,
|
|
"num_tokens": 525884.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.081359624862671,
|
|
"epoch": 0.013965388181929906,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5558,
|
|
"mean_token_accuracy": 0.145179907977581,
|
|
"num_tokens": 530373.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.055321979522705,
|
|
"epoch": 0.014077111287385344,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3843,
|
|
"mean_token_accuracy": 0.15800822898745537,
|
|
"num_tokens": 534571.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.117716646194458,
|
|
"epoch": 0.014188834392840783,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5609,
|
|
"mean_token_accuracy": 0.13290601670742036,
|
|
"num_tokens": 538938.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.100346279144287,
|
|
"epoch": 0.014300557498296222,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5165,
|
|
"mean_token_accuracy": 0.1455024905502796,
|
|
"num_tokens": 542977.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.071889972686767,
|
|
"epoch": 0.014412280603751661,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4884,
|
|
"mean_token_accuracy": 0.1465342827141285,
|
|
"num_tokens": 547005.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.0628125190734865,
|
|
"epoch": 0.014524003709207102,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5272,
|
|
"mean_token_accuracy": 0.13975051417946815,
|
|
"num_tokens": 551361.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.023661518096924,
|
|
"epoch": 0.01463572681466254,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4833,
|
|
"mean_token_accuracy": 0.14340668320655822,
|
|
"num_tokens": 555574.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.989748096466064,
|
|
"epoch": 0.01474744992011798,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5127,
|
|
"mean_token_accuracy": 0.14289377629756927,
|
|
"num_tokens": 560368.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.077346086502075,
|
|
"epoch": 0.014859173025573418,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5634,
|
|
"mean_token_accuracy": 0.1373551793396473,
|
|
"num_tokens": 564884.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.1274079322814945,
|
|
"epoch": 0.014970896131028859,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5586,
|
|
"mean_token_accuracy": 0.1458572693169117,
|
|
"num_tokens": 569494.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.002107572555542,
|
|
"epoch": 0.015082619236484298,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5547,
|
|
"mean_token_accuracy": 0.13712208420038224,
|
|
"num_tokens": 574161.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.072530221939087,
|
|
"epoch": 0.015194342341939737,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4626,
|
|
"mean_token_accuracy": 0.14953978583216668,
|
|
"num_tokens": 578510.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.912496089935303,
|
|
"epoch": 0.015306065447395175,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4979,
|
|
"mean_token_accuracy": 0.1424515500664711,
|
|
"num_tokens": 582695.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.025566148757934,
|
|
"epoch": 0.015417788552850614,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4376,
|
|
"mean_token_accuracy": 0.13332833126187324,
|
|
"num_tokens": 586622.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.013347387313843,
|
|
"epoch": 0.015529511658306055,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4683,
|
|
"mean_token_accuracy": 0.14548797011375428,
|
|
"num_tokens": 591704.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.755953073501587,
|
|
"epoch": 0.015641234763761492,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.276,
|
|
"mean_token_accuracy": 0.15882887542247773,
|
|
"num_tokens": 596029.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.987126111984253,
|
|
"epoch": 0.015752957869216933,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4419,
|
|
"mean_token_accuracy": 0.14538582488894464,
|
|
"num_tokens": 600115.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.84918212890625,
|
|
"epoch": 0.015864680974672373,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.4459,
|
|
"mean_token_accuracy": 0.14390605613589286,
|
|
"num_tokens": 604488.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.084057378768921,
|
|
"epoch": 0.01597640408012781,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4439,
|
|
"mean_token_accuracy": 0.14581422209739686,
|
|
"num_tokens": 608776.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.7030833721160885,
|
|
"epoch": 0.01608812718558325,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4504,
|
|
"mean_token_accuracy": 0.15099047794938086,
|
|
"num_tokens": 612771.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.734662055969238,
|
|
"epoch": 0.01619985029103869,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3566,
|
|
"mean_token_accuracy": 0.1516010656952858,
|
|
"num_tokens": 616947.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.9721879959106445,
|
|
"epoch": 0.01631157339649413,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4281,
|
|
"mean_token_accuracy": 0.15130600407719613,
|
|
"num_tokens": 621064.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.856808233261108,
|
|
"epoch": 0.01642329650194957,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.4363,
|
|
"mean_token_accuracy": 0.14606723934412003,
|
|
"num_tokens": 625349.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.863543367385864,
|
|
"epoch": 0.016535019607405006,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3498,
|
|
"mean_token_accuracy": 0.15212762504816055,
|
|
"num_tokens": 629754.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.9013481616973875,
|
|
"epoch": 0.016646742712860447,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3994,
|
|
"mean_token_accuracy": 0.13600233122706412,
|
|
"num_tokens": 634111.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.783719491958618,
|
|
"epoch": 0.016758465818315887,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2659,
|
|
"mean_token_accuracy": 0.16154912412166594,
|
|
"num_tokens": 638527.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.681515502929687,
|
|
"epoch": 0.016870188923771325,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2959,
|
|
"mean_token_accuracy": 0.15193597078323365,
|
|
"num_tokens": 642530.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.832862663269043,
|
|
"epoch": 0.016981912029226765,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.15937515050172807,
|
|
"num_tokens": 646707.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.755436944961548,
|
|
"epoch": 0.017093635134682202,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4113,
|
|
"mean_token_accuracy": 0.15184309035539628,
|
|
"num_tokens": 651326.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.871073818206787,
|
|
"epoch": 0.017205358240137643,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4618,
|
|
"mean_token_accuracy": 0.15124865621328354,
|
|
"num_tokens": 655651.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.727626895904541,
|
|
"epoch": 0.017317081345593083,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3946,
|
|
"mean_token_accuracy": 0.14350455030798911,
|
|
"num_tokens": 659868.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.921767711639404,
|
|
"epoch": 0.01742880445104852,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4403,
|
|
"mean_token_accuracy": 0.14834593906998633,
|
|
"num_tokens": 664404.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.699166393280029,
|
|
"epoch": 0.01754052755650396,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2549,
|
|
"mean_token_accuracy": 0.1540958382189274,
|
|
"num_tokens": 668543.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.646093511581421,
|
|
"epoch": 0.0176522506619594,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2322,
|
|
"mean_token_accuracy": 0.15864449143409728,
|
|
"num_tokens": 672637.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.7683539390563965,
|
|
"epoch": 0.01776397376741484,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.3473,
|
|
"mean_token_accuracy": 0.15486485213041307,
|
|
"num_tokens": 676929.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.6748552322387695,
|
|
"epoch": 0.01787569687287028,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3935,
|
|
"mean_token_accuracy": 0.14576203897595405,
|
|
"num_tokens": 681287.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.773136854171753,
|
|
"epoch": 0.017987419978325717,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.2896,
|
|
"mean_token_accuracy": 0.14774783104658126,
|
|
"num_tokens": 685543.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.736042118072509,
|
|
"epoch": 0.018099143083781157,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2738,
|
|
"mean_token_accuracy": 0.14571748450398445,
|
|
"num_tokens": 689479.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.654993534088135,
|
|
"epoch": 0.018210866189236594,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2274,
|
|
"mean_token_accuracy": 0.16590481102466584,
|
|
"num_tokens": 693564.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.655237770080566,
|
|
"epoch": 0.018322589294692035,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3025,
|
|
"mean_token_accuracy": 0.15087175816297532,
|
|
"num_tokens": 697719.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.695311594009399,
|
|
"epoch": 0.018434312400147475,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.1513037145137787,
|
|
"num_tokens": 701937.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.683870124816894,
|
|
"epoch": 0.018546035505602913,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3653,
|
|
"mean_token_accuracy": 0.14852157458662987,
|
|
"num_tokens": 706457.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.849027347564697,
|
|
"epoch": 0.018657758611058353,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.38,
|
|
"mean_token_accuracy": 0.15221845954656602,
|
|
"num_tokens": 710538.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.5683678150177,
|
|
"epoch": 0.018769481716513794,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.248,
|
|
"mean_token_accuracy": 0.16236387193202972,
|
|
"num_tokens": 714663.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.697820091247559,
|
|
"epoch": 0.01888120482196923,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2517,
|
|
"mean_token_accuracy": 0.1534279391169548,
|
|
"num_tokens": 718828.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.571414041519165,
|
|
"epoch": 0.01899292792742467,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.2619,
|
|
"mean_token_accuracy": 0.15856588035821914,
|
|
"num_tokens": 723155.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.808944749832153,
|
|
"epoch": 0.01910465103288011,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.3217,
|
|
"mean_token_accuracy": 0.15492385476827622,
|
|
"num_tokens": 727605.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.5332495212554935,
|
|
"epoch": 0.01921637413833555,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.2777,
|
|
"mean_token_accuracy": 0.14969536513090134,
|
|
"num_tokens": 731510.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.71437783241272,
|
|
"epoch": 0.01932809724379099,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4619,
|
|
"mean_token_accuracy": 0.1470765456557274,
|
|
"num_tokens": 736048.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.678787279129028,
|
|
"epoch": 0.019439820349246427,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2512,
|
|
"mean_token_accuracy": 0.15854543596506118,
|
|
"num_tokens": 740714.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.619452238082886,
|
|
"epoch": 0.019551543454701868,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.3257,
|
|
"mean_token_accuracy": 0.15650657266378404,
|
|
"num_tokens": 745167.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.694642496109009,
|
|
"epoch": 0.019663266560157305,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1537,
|
|
"mean_token_accuracy": 0.16280549690127372,
|
|
"num_tokens": 749401.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.627922868728637,
|
|
"epoch": 0.019774989665612745,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2224,
|
|
"mean_token_accuracy": 0.16411646455526352,
|
|
"num_tokens": 753554.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.636331701278687,
|
|
"epoch": 0.019886712771068186,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2725,
|
|
"mean_token_accuracy": 0.14831542521715163,
|
|
"num_tokens": 757744.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.588339996337891,
|
|
"epoch": 0.019998435876523623,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1633,
|
|
"mean_token_accuracy": 0.15396574288606643,
|
|
"num_tokens": 761858.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.522528743743896,
|
|
"epoch": 0.020110158981979064,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.1908,
|
|
"mean_token_accuracy": 0.16618741899728776,
|
|
"num_tokens": 765852.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.504194116592407,
|
|
"epoch": 0.0202218820874345,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.195,
|
|
"mean_token_accuracy": 0.1562537170946598,
|
|
"num_tokens": 770274.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.654917287826538,
|
|
"epoch": 0.02033360519288994,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2299,
|
|
"mean_token_accuracy": 0.15019148588180542,
|
|
"num_tokens": 774775.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.611723136901856,
|
|
"epoch": 0.020445328298345382,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2769,
|
|
"mean_token_accuracy": 0.16535960435867308,
|
|
"num_tokens": 779374.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.6483289241790775,
|
|
"epoch": 0.02055705140380082,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2519,
|
|
"mean_token_accuracy": 0.1526936858892441,
|
|
"num_tokens": 783537.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.512672090530396,
|
|
"epoch": 0.02066877450925626,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3001,
|
|
"mean_token_accuracy": 0.15667854249477386,
|
|
"num_tokens": 787562.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.7299144744873045,
|
|
"epoch": 0.0207804976147117,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3687,
|
|
"mean_token_accuracy": 0.1443271040916443,
|
|
"num_tokens": 792303.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.5588274002075195,
|
|
"epoch": 0.020892220720167137,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1584,
|
|
"mean_token_accuracy": 0.16573118567466735,
|
|
"num_tokens": 796402.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.613452911376953,
|
|
"epoch": 0.021003943825622578,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2314,
|
|
"mean_token_accuracy": 0.16508372873067856,
|
|
"num_tokens": 800004.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.341346979141235,
|
|
"epoch": 0.021115666931078015,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.117,
|
|
"mean_token_accuracy": 0.16131858825683593,
|
|
"num_tokens": 803782.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.510941028594971,
|
|
"epoch": 0.021227390036533456,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1673,
|
|
"mean_token_accuracy": 0.16098449528217315,
|
|
"num_tokens": 808013.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.4510125637054445,
|
|
"epoch": 0.021339113141988896,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1669,
|
|
"mean_token_accuracy": 0.1604609191417694,
|
|
"num_tokens": 812130.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.501435708999634,
|
|
"epoch": 0.021450836247444333,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1342,
|
|
"mean_token_accuracy": 0.16159728765487671,
|
|
"num_tokens": 816246.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.510243511199951,
|
|
"epoch": 0.021562559352899774,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1317,
|
|
"mean_token_accuracy": 0.16103896945714952,
|
|
"num_tokens": 820628.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.387259864807129,
|
|
"epoch": 0.02167428245835521,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1367,
|
|
"mean_token_accuracy": 0.1684929057955742,
|
|
"num_tokens": 824864.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.61003623008728,
|
|
"epoch": 0.02178600556381065,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.183,
|
|
"mean_token_accuracy": 0.15778308734297752,
|
|
"num_tokens": 828887.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.448976230621338,
|
|
"epoch": 0.021897728669266092,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0739,
|
|
"mean_token_accuracy": 0.16717422604560853,
|
|
"num_tokens": 832784.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.41447229385376,
|
|
"epoch": 0.02200945177472153,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0937,
|
|
"mean_token_accuracy": 0.15710036903619767,
|
|
"num_tokens": 837408.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.451285934448242,
|
|
"epoch": 0.02212117488017697,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1512,
|
|
"mean_token_accuracy": 0.16300837993621825,
|
|
"num_tokens": 841536.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.426507759094238,
|
|
"epoch": 0.022232897985632407,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0582,
|
|
"mean_token_accuracy": 0.17121603935956956,
|
|
"num_tokens": 846044.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.320924186706543,
|
|
"epoch": 0.022344621091087848,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1118,
|
|
"mean_token_accuracy": 0.1685657724738121,
|
|
"num_tokens": 850254.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.453368425369263,
|
|
"epoch": 0.022456344196543288,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.0715,
|
|
"mean_token_accuracy": 0.16939653158187867,
|
|
"num_tokens": 854548.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.386558151245117,
|
|
"epoch": 0.022568067301998725,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.2133,
|
|
"mean_token_accuracy": 0.1603351503610611,
|
|
"num_tokens": 858837.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.328588962554932,
|
|
"epoch": 0.022679790407454166,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.0295,
|
|
"mean_token_accuracy": 0.16671301424503326,
|
|
"num_tokens": 862772.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.418800020217896,
|
|
"epoch": 0.022791513512909607,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.1171,
|
|
"mean_token_accuracy": 0.1632360801100731,
|
|
"num_tokens": 866909.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.417377662658692,
|
|
"epoch": 0.022903236618365044,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.2569,
|
|
"mean_token_accuracy": 0.1513159267604351,
|
|
"num_tokens": 871127.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.434281063079834,
|
|
"epoch": 0.023014959723820484,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.1777,
|
|
"mean_token_accuracy": 0.1645081579685211,
|
|
"num_tokens": 875292.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.387321043014526,
|
|
"epoch": 0.02312668282927592,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.15736222565174102,
|
|
"num_tokens": 879286.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.443514776229859,
|
|
"epoch": 0.023238405934731362,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.1128,
|
|
"mean_token_accuracy": 0.16863536983728408,
|
|
"num_tokens": 883490.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.404753255844116,
|
|
"epoch": 0.023350129040186803,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.0757,
|
|
"mean_token_accuracy": 0.1634724885225296,
|
|
"num_tokens": 887786.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.409213829040527,
|
|
"epoch": 0.02346185214564224,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.0974,
|
|
"mean_token_accuracy": 0.16956386119127273,
|
|
"num_tokens": 892035.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.431076288223267,
|
|
"epoch": 0.02357357525109768,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.0877,
|
|
"mean_token_accuracy": 0.16654614359140396,
|
|
"num_tokens": 896099.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.382920742034912,
|
|
"epoch": 0.023685298356553117,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.0172,
|
|
"mean_token_accuracy": 0.17494994923472404,
|
|
"num_tokens": 900448.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.291054439544678,
|
|
"epoch": 0.023797021462008558,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.0807,
|
|
"mean_token_accuracy": 0.16231946498155594,
|
|
"num_tokens": 904538.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.47989239692688,
|
|
"epoch": 0.023908744567464,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.1925,
|
|
"mean_token_accuracy": 0.15721007734537124,
|
|
"num_tokens": 908680.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.449333000183105,
|
|
"epoch": 0.024020467672919436,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.15536463409662246,
|
|
"num_tokens": 912868.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.176778936386109,
|
|
"epoch": 0.024132190778374876,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.1923,
|
|
"mean_token_accuracy": 0.15709846168756486,
|
|
"num_tokens": 917312.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.612870168685913,
|
|
"epoch": 0.024243913883830313,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1957,
|
|
"mean_token_accuracy": 0.16389840096235275,
|
|
"num_tokens": 921977.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.3305353164672855,
|
|
"epoch": 0.024355636989285754,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.1261,
|
|
"mean_token_accuracy": 0.17108615338802338,
|
|
"num_tokens": 925964.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.32656307220459,
|
|
"epoch": 0.024467360094741195,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 5.9851,
|
|
"mean_token_accuracy": 0.1709235593676567,
|
|
"num_tokens": 930077.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.294446897506714,
|
|
"epoch": 0.02457908320019663,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.0255,
|
|
"mean_token_accuracy": 0.17414780110120773,
|
|
"num_tokens": 934547.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.341350984573364,
|
|
"epoch": 0.024690806305652072,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.0495,
|
|
"mean_token_accuracy": 0.17528252005577089,
|
|
"num_tokens": 938583.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.444618558883667,
|
|
"epoch": 0.024802529411107513,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.0981,
|
|
"mean_token_accuracy": 0.16719860881567,
|
|
"num_tokens": 943007.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.317926597595215,
|
|
"epoch": 0.02491425251656295,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.0836,
|
|
"mean_token_accuracy": 0.16416927874088288,
|
|
"num_tokens": 947281.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.2807488441467285,
|
|
"epoch": 0.02502597562201839,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 5.9665,
|
|
"mean_token_accuracy": 0.1739448979496956,
|
|
"num_tokens": 951443.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.281340885162353,
|
|
"epoch": 0.025137698727473828,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.068,
|
|
"mean_token_accuracy": 0.15851529389619828,
|
|
"num_tokens": 955131.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.28678936958313,
|
|
"epoch": 0.02524942183292927,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.0043,
|
|
"mean_token_accuracy": 0.17229326516389848,
|
|
"num_tokens": 959408.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.292270088195801,
|
|
"epoch": 0.02536114493838471,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.067,
|
|
"mean_token_accuracy": 0.1669071465730667,
|
|
"num_tokens": 963657.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.207029819488525,
|
|
"epoch": 0.025472868043840146,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.0802,
|
|
"mean_token_accuracy": 0.16384815871715547,
|
|
"num_tokens": 967973.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.254129791259766,
|
|
"epoch": 0.025584591149295587,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 5.8691,
|
|
"mean_token_accuracy": 0.1844624474644661,
|
|
"num_tokens": 971957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.198523759841919,
|
|
"epoch": 0.025696314254751024,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 5.9141,
|
|
"mean_token_accuracy": 0.1774628072977066,
|
|
"num_tokens": 975679.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.320246267318725,
|
|
"epoch": 0.025808037360206464,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.1461,
|
|
"mean_token_accuracy": 0.16569039970636368,
|
|
"num_tokens": 979806.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.319731616973877,
|
|
"epoch": 0.025919760465661905,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.0224,
|
|
"mean_token_accuracy": 0.17389359325170517,
|
|
"num_tokens": 984221.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.255821323394775,
|
|
"epoch": 0.026031483571117342,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 5.9962,
|
|
"mean_token_accuracy": 0.16791471540927888,
|
|
"num_tokens": 988323.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.295364618301392,
|
|
"epoch": 0.026143206676572783,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.0453,
|
|
"mean_token_accuracy": 0.16588685512542725,
|
|
"num_tokens": 992643.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.16099534034729,
|
|
"epoch": 0.02625492978202822,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 5.8792,
|
|
"mean_token_accuracy": 0.18837386816740037,
|
|
"num_tokens": 996994.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.207149171829224,
|
|
"epoch": 0.02636665288748366,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.0087,
|
|
"mean_token_accuracy": 0.17770088016986846,
|
|
"num_tokens": 1001345.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.242070484161377,
|
|
"epoch": 0.0264783759929391,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 5.9216,
|
|
"mean_token_accuracy": 0.17433482706546782,
|
|
"num_tokens": 1005839.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.214767789840698,
|
|
"epoch": 0.026590099098394538,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.16556215435266494,
|
|
"num_tokens": 1010698.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.351368093490601,
|
|
"epoch": 0.02670182220384998,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.029,
|
|
"mean_token_accuracy": 0.17102244794368743,
|
|
"num_tokens": 1014982.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.117547225952149,
|
|
"epoch": 0.026813545309305416,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 5.9538,
|
|
"mean_token_accuracy": 0.1834021046757698,
|
|
"num_tokens": 1019235.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.2841479778289795,
|
|
"epoch": 0.026925268414760856,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.037,
|
|
"mean_token_accuracy": 0.17534659653902054,
|
|
"num_tokens": 1023855.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.276741075515747,
|
|
"epoch": 0.027036991520216297,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 5.9946,
|
|
"mean_token_accuracy": 0.179076386988163,
|
|
"num_tokens": 1028464.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.3194655418396,
|
|
"epoch": 0.027148714625671734,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 5.9019,
|
|
"mean_token_accuracy": 0.17992053031921387,
|
|
"num_tokens": 1033110.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.059362745285034,
|
|
"epoch": 0.027260437731127175,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.0265,
|
|
"mean_token_accuracy": 0.16747127920389177,
|
|
"num_tokens": 1037470.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.275005626678467,
|
|
"epoch": 0.027372160836582615,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 5.9895,
|
|
"mean_token_accuracy": 0.1710431009531021,
|
|
"num_tokens": 1041662.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.280491781234741,
|
|
"epoch": 0.027483883942038052,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.17440603524446488,
|
|
"num_tokens": 1046041.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.002849531173706,
|
|
"epoch": 0.027595607047493493,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 5.9419,
|
|
"mean_token_accuracy": 0.17400026842951774,
|
|
"num_tokens": 1049893.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.270329856872559,
|
|
"epoch": 0.02770733015294893,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.0921,
|
|
"mean_token_accuracy": 0.16643197387456893,
|
|
"num_tokens": 1054048.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.235016345977783,
|
|
"epoch": 0.02781905325840437,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.16786330044269562,
|
|
"num_tokens": 1058176.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.242469263076782,
|
|
"epoch": 0.02793077636385981,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 5.9481,
|
|
"mean_token_accuracy": 0.16720382124185562,
|
|
"num_tokens": 1062031.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.049459505081177,
|
|
"epoch": 0.02804249946931525,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 5.9364,
|
|
"mean_token_accuracy": 0.1834051489830017,
|
|
"num_tokens": 1066705.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.242766857147217,
|
|
"epoch": 0.02815422257477069,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 5.9192,
|
|
"mean_token_accuracy": 0.18238281905651094,
|
|
"num_tokens": 1070960.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.3449784278869625,
|
|
"epoch": 0.028265945680226126,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.1001,
|
|
"mean_token_accuracy": 0.16833141073584557,
|
|
"num_tokens": 1075576.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.116433477401733,
|
|
"epoch": 0.028377668785681567,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.17545292675495147,
|
|
"num_tokens": 1079538.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.0287150859832765,
|
|
"epoch": 0.028489391891137007,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 5.9247,
|
|
"mean_token_accuracy": 0.17888442128896714,
|
|
"num_tokens": 1083296.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.130237722396851,
|
|
"epoch": 0.028601114996592444,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 5.9678,
|
|
"mean_token_accuracy": 0.17468917220830918,
|
|
"num_tokens": 1087688.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.155571603775025,
|
|
"epoch": 0.028712838102047885,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 5.9098,
|
|
"mean_token_accuracy": 0.1857583463191986,
|
|
"num_tokens": 1091988.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.23262209892273,
|
|
"epoch": 0.028824561207503322,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.17543941140174865,
|
|
"num_tokens": 1095949.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.130452013015747,
|
|
"epoch": 0.028936284312958763,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 5.8586,
|
|
"mean_token_accuracy": 0.17781955897808074,
|
|
"num_tokens": 1099795.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.954253005981445,
|
|
"epoch": 0.029048007418414203,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 5.7856,
|
|
"mean_token_accuracy": 0.1847798377275467,
|
|
"num_tokens": 1103894.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.213930320739746,
|
|
"epoch": 0.02915973052386964,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.0079,
|
|
"mean_token_accuracy": 0.17599031776189805,
|
|
"num_tokens": 1108382.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.095987224578858,
|
|
"epoch": 0.02927145362932508,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 5.7812,
|
|
"mean_token_accuracy": 0.18635666817426683,
|
|
"num_tokens": 1112841.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.131918239593506,
|
|
"epoch": 0.02938317673478052,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.0198,
|
|
"mean_token_accuracy": 0.17015290707349778,
|
|
"num_tokens": 1117319.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.14964075088501,
|
|
"epoch": 0.02949489984023596,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 5.9832,
|
|
"mean_token_accuracy": 0.17692959010601045,
|
|
"num_tokens": 1121371.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.087015199661255,
|
|
"epoch": 0.0296066229456914,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.1832718223333359,
|
|
"num_tokens": 1125496.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.156226062774659,
|
|
"epoch": 0.029718346051146836,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.18602917045354844,
|
|
"num_tokens": 1129929.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.084760046005249,
|
|
"epoch": 0.029830069156602277,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 5.8466,
|
|
"mean_token_accuracy": 0.18009257912635804,
|
|
"num_tokens": 1133982.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.020449161529541,
|
|
"epoch": 0.029941792262057718,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.176700659096241,
|
|
"num_tokens": 1138270.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.241879224777222,
|
|
"epoch": 0.030053515367513155,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 5.9128,
|
|
"mean_token_accuracy": 0.1819134071469307,
|
|
"num_tokens": 1142197.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.1059808254241945,
|
|
"epoch": 0.030165238472968595,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 5.9763,
|
|
"mean_token_accuracy": 0.17749614864587784,
|
|
"num_tokens": 1146439.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.060887145996094,
|
|
"epoch": 0.030276961578424032,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 5.9363,
|
|
"mean_token_accuracy": 0.1883004993200302,
|
|
"num_tokens": 1150809.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.27038950920105,
|
|
"epoch": 0.030388684683879473,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.0562,
|
|
"mean_token_accuracy": 0.1610119305551052,
|
|
"num_tokens": 1155132.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.380424118041992,
|
|
"epoch": 0.030500407789334914,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 5.9975,
|
|
"mean_token_accuracy": 0.167344368994236,
|
|
"num_tokens": 1159694.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.135957479476929,
|
|
"epoch": 0.03061213089479035,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 5.9235,
|
|
"mean_token_accuracy": 0.1690131738781929,
|
|
"num_tokens": 1164070.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.074416303634644,
|
|
"epoch": 0.03072385400024579,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 5.7706,
|
|
"mean_token_accuracy": 0.19714766144752502,
|
|
"num_tokens": 1168097.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.055906534194946,
|
|
"epoch": 0.03083557710570123,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 5.9315,
|
|
"mean_token_accuracy": 0.1687031865119934,
|
|
"num_tokens": 1172375.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.07122950553894,
|
|
"epoch": 0.03094730021115667,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 5.962,
|
|
"mean_token_accuracy": 0.16933582127094268,
|
|
"num_tokens": 1176628.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.064983272552491,
|
|
"epoch": 0.03105902331661211,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 5.9208,
|
|
"mean_token_accuracy": 0.1758138954639435,
|
|
"num_tokens": 1180500.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.065640687942505,
|
|
"epoch": 0.031170746422067547,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 5.9367,
|
|
"mean_token_accuracy": 0.17107723653316498,
|
|
"num_tokens": 1184883.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.242153358459473,
|
|
"epoch": 0.031282469527522984,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.17483978867530822,
|
|
"num_tokens": 1188723.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.178732633590698,
|
|
"epoch": 0.03139419263297843,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.0475,
|
|
"mean_token_accuracy": 0.1655319780111313,
|
|
"num_tokens": 1193318.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.136021947860717,
|
|
"epoch": 0.031505915738433865,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.17083597481250762,
|
|
"num_tokens": 1197663.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.984335994720459,
|
|
"epoch": 0.0316176388438893,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 5.7723,
|
|
"mean_token_accuracy": 0.19360799193382264,
|
|
"num_tokens": 1201949.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.23677077293396,
|
|
"epoch": 0.031729361949344746,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.0375,
|
|
"mean_token_accuracy": 0.1644550994038582,
|
|
"num_tokens": 1206360.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.234195375442505,
|
|
"epoch": 0.03184108505480018,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.17676099985837937,
|
|
"num_tokens": 1210738.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.093136215209961,
|
|
"epoch": 0.03195280816025562,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 5.8809,
|
|
"mean_token_accuracy": 0.17498217523097992,
|
|
"num_tokens": 1214722.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.140736865997314,
|
|
"epoch": 0.032064531265711065,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 5.9662,
|
|
"mean_token_accuracy": 0.17408566623926164,
|
|
"num_tokens": 1218811.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.182424402236938,
|
|
"epoch": 0.0321762543711665,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 5.9361,
|
|
"mean_token_accuracy": 0.17186458110809327,
|
|
"num_tokens": 1223089.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.910596895217895,
|
|
"epoch": 0.03228797747662194,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 5.7536,
|
|
"mean_token_accuracy": 0.18107606768608092,
|
|
"num_tokens": 1226970.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.991322135925293,
|
|
"epoch": 0.03239970058207738,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 5.8273,
|
|
"mean_token_accuracy": 0.18494699150323868,
|
|
"num_tokens": 1230734.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.019584083557129,
|
|
"epoch": 0.03251142368753282,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 5.887,
|
|
"mean_token_accuracy": 0.1700140006840229,
|
|
"num_tokens": 1235363.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.014257526397705,
|
|
"epoch": 0.03262314679298826,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 5.7371,
|
|
"mean_token_accuracy": 0.18364810347557067,
|
|
"num_tokens": 1239826.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.125292587280273,
|
|
"epoch": 0.032734869898443694,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 5.9677,
|
|
"mean_token_accuracy": 0.16234042197465898,
|
|
"num_tokens": 1244067.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.030685091018677,
|
|
"epoch": 0.03284659300389914,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 5.7734,
|
|
"mean_token_accuracy": 0.18260960876941681,
|
|
"num_tokens": 1248307.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.964048194885254,
|
|
"epoch": 0.032958316109354575,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.17982411235570908,
|
|
"num_tokens": 1252504.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.956363296508789,
|
|
"epoch": 0.03307003921481001,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 5.8804,
|
|
"mean_token_accuracy": 0.1755678966641426,
|
|
"num_tokens": 1256823.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.03358826637268,
|
|
"epoch": 0.03318176232026546,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 5.8242,
|
|
"mean_token_accuracy": 0.17837631106376647,
|
|
"num_tokens": 1261236.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.114140462875366,
|
|
"epoch": 0.033293485425720894,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.17883239686489105,
|
|
"num_tokens": 1265360.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.907846260070801,
|
|
"epoch": 0.03340520853117633,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 5.7961,
|
|
"mean_token_accuracy": 0.1842688113451004,
|
|
"num_tokens": 1269683.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.970038223266601,
|
|
"epoch": 0.033516931636631775,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 5.8931,
|
|
"mean_token_accuracy": 0.1749550759792328,
|
|
"num_tokens": 1273843.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.059222936630249,
|
|
"epoch": 0.03362865474208721,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.182131527364254,
|
|
"num_tokens": 1278299.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.9348760604858395,
|
|
"epoch": 0.03374037784754265,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 5.7727,
|
|
"mean_token_accuracy": 0.18791095465421676,
|
|
"num_tokens": 1282541.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.046141624450684,
|
|
"epoch": 0.033852100952998086,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 5.6971,
|
|
"mean_token_accuracy": 0.18971533626317977,
|
|
"num_tokens": 1286318.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.891553258895874,
|
|
"epoch": 0.03396382405845353,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.18425178676843643,
|
|
"num_tokens": 1290713.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.0908989906311035,
|
|
"epoch": 0.03407554716390897,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 5.7639,
|
|
"mean_token_accuracy": 0.17793523371219636,
|
|
"num_tokens": 1294820.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.103552007675171,
|
|
"epoch": 0.034187270269364405,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 5.7944,
|
|
"mean_token_accuracy": 0.1779856227338314,
|
|
"num_tokens": 1299232.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.96611704826355,
|
|
"epoch": 0.03429899337481985,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 5.8039,
|
|
"mean_token_accuracy": 0.18248388022184373,
|
|
"num_tokens": 1303287.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.9684473991394045,
|
|
"epoch": 0.034410716480275286,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 5.778,
|
|
"mean_token_accuracy": 0.18929605185985565,
|
|
"num_tokens": 1307563.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.80886549949646,
|
|
"epoch": 0.03452243958573072,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 5.6779,
|
|
"mean_token_accuracy": 0.18525782525539397,
|
|
"num_tokens": 1311484.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.063689661026001,
|
|
"epoch": 0.03463416269118617,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 5.9013,
|
|
"mean_token_accuracy": 0.17240882217884063,
|
|
"num_tokens": 1316093.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.139868497848511,
|
|
"epoch": 0.034745885796641604,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 5.8746,
|
|
"mean_token_accuracy": 0.1783792719244957,
|
|
"num_tokens": 1321001.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.060155868530273,
|
|
"epoch": 0.03485760890209704,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 5.7142,
|
|
"mean_token_accuracy": 0.18678450286388398,
|
|
"num_tokens": 1325069.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.023049211502075,
|
|
"epoch": 0.034969332007552485,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 5.6733,
|
|
"mean_token_accuracy": 0.18095604628324508,
|
|
"num_tokens": 1328887.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.019297122955322,
|
|
"epoch": 0.03508105511300792,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 5.9267,
|
|
"mean_token_accuracy": 0.17412633448839188,
|
|
"num_tokens": 1333395.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.182040786743164,
|
|
"epoch": 0.03519277821846336,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.0356,
|
|
"mean_token_accuracy": 0.16573426127433777,
|
|
"num_tokens": 1337457.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.961935520172119,
|
|
"epoch": 0.0353045013239188,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 5.6475,
|
|
"mean_token_accuracy": 0.1947586268186569,
|
|
"num_tokens": 1341768.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.973260307312012,
|
|
"epoch": 0.03541622442937424,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 5.7183,
|
|
"mean_token_accuracy": 0.1928910493850708,
|
|
"num_tokens": 1346191.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.966824626922607,
|
|
"epoch": 0.03552794753482968,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 5.7661,
|
|
"mean_token_accuracy": 0.18546913117170333,
|
|
"num_tokens": 1350228.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.945509767532348,
|
|
"epoch": 0.035639670640285115,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 5.8698,
|
|
"mean_token_accuracy": 0.17462340891361236,
|
|
"num_tokens": 1354722.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.016447973251343,
|
|
"epoch": 0.03575139374574056,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 5.863,
|
|
"mean_token_accuracy": 0.18090844303369522,
|
|
"num_tokens": 1358985.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.129895305633545,
|
|
"epoch": 0.035863116851195996,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 5.7705,
|
|
"mean_token_accuracy": 0.18566034287214278,
|
|
"num_tokens": 1363281.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.03486967086792,
|
|
"epoch": 0.03597483995665143,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 5.8546,
|
|
"mean_token_accuracy": 0.17603144496679307,
|
|
"num_tokens": 1367777.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.92125449180603,
|
|
"epoch": 0.03608656306210688,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 5.9066,
|
|
"mean_token_accuracy": 0.16197986379265786,
|
|
"num_tokens": 1372537.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.871002578735352,
|
|
"epoch": 0.036198286167562314,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 5.6652,
|
|
"mean_token_accuracy": 0.19525430649518966,
|
|
"num_tokens": 1377341.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.010153532028198,
|
|
"epoch": 0.03631000927301775,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 5.7759,
|
|
"mean_token_accuracy": 0.17927465140819548,
|
|
"num_tokens": 1381480.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.026579904556274,
|
|
"epoch": 0.03642173237847319,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 5.7237,
|
|
"mean_token_accuracy": 0.17483969032764435,
|
|
"num_tokens": 1386057.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.877592515945435,
|
|
"epoch": 0.03653345548392863,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 5.7144,
|
|
"mean_token_accuracy": 0.18803704977035524,
|
|
"num_tokens": 1390405.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.916935634613037,
|
|
"epoch": 0.03664517858938407,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 5.7084,
|
|
"mean_token_accuracy": 0.18442935198545457,
|
|
"num_tokens": 1394972.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.940288639068603,
|
|
"epoch": 0.03675690169483951,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 5.687,
|
|
"mean_token_accuracy": 0.19444524943828584,
|
|
"num_tokens": 1399270.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.900350522994995,
|
|
"epoch": 0.03686862480029495,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 5.686,
|
|
"mean_token_accuracy": 0.19073100686073302,
|
|
"num_tokens": 1403506.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.816886520385742,
|
|
"epoch": 0.03698034790575039,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 5.8022,
|
|
"mean_token_accuracy": 0.18439093232154846,
|
|
"num_tokens": 1407493.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.997721147537232,
|
|
"epoch": 0.037092071011205825,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 5.7482,
|
|
"mean_token_accuracy": 0.1847753569483757,
|
|
"num_tokens": 1411871.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.972122716903686,
|
|
"epoch": 0.03720379411666127,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 5.7221,
|
|
"mean_token_accuracy": 0.18191029131412506,
|
|
"num_tokens": 1415951.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.838072156906128,
|
|
"epoch": 0.037315517222116706,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 5.7248,
|
|
"mean_token_accuracy": 0.18715287148952484,
|
|
"num_tokens": 1420394.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.02448787689209,
|
|
"epoch": 0.037427240327572144,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 5.8614,
|
|
"mean_token_accuracy": 0.1815338611602783,
|
|
"num_tokens": 1424921.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.867510223388672,
|
|
"epoch": 0.03753896343302759,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 5.6933,
|
|
"mean_token_accuracy": 0.1878269910812378,
|
|
"num_tokens": 1428638.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.935184717178345,
|
|
"epoch": 0.037650686538483025,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.7843,
|
|
"mean_token_accuracy": 0.18289850652217865,
|
|
"num_tokens": 1432780.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.906549501419067,
|
|
"epoch": 0.03776240964393846,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 5.7056,
|
|
"mean_token_accuracy": 0.18908923119306564,
|
|
"num_tokens": 1437178.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.072110605239868,
|
|
"epoch": 0.0378741327493939,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 5.7695,
|
|
"mean_token_accuracy": 0.17374685257673264,
|
|
"num_tokens": 1441040.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.934421253204346,
|
|
"epoch": 0.03798585585484934,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 5.7489,
|
|
"mean_token_accuracy": 0.18246821612119674,
|
|
"num_tokens": 1445530.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.934323310852051,
|
|
"epoch": 0.03809757896030478,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.18564676642417907,
|
|
"num_tokens": 1449623.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.9197345733642575,
|
|
"epoch": 0.03820930206576022,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 5.6831,
|
|
"mean_token_accuracy": 0.18841421008110046,
|
|
"num_tokens": 1454167.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.848270463943481,
|
|
"epoch": 0.03832102517121566,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 5.6999,
|
|
"mean_token_accuracy": 0.18317352831363679,
|
|
"num_tokens": 1458666.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.861880540847778,
|
|
"epoch": 0.0384327482766711,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 5.7067,
|
|
"mean_token_accuracy": 0.19191939234733582,
|
|
"num_tokens": 1463253.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.804866981506348,
|
|
"epoch": 0.038544471382126536,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 5.6554,
|
|
"mean_token_accuracy": 0.18410271555185317,
|
|
"num_tokens": 1467487.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.91278042793274,
|
|
"epoch": 0.03865619448758198,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.1775332771241665,
|
|
"num_tokens": 1471408.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.8655084609985355,
|
|
"epoch": 0.03876791759303742,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.20361874550580977,
|
|
"num_tokens": 1475296.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.9305091381073,
|
|
"epoch": 0.038879640698492854,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 5.7256,
|
|
"mean_token_accuracy": 0.1764163628220558,
|
|
"num_tokens": 1479593.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.91884412765503,
|
|
"epoch": 0.0389913638039483,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 5.6551,
|
|
"mean_token_accuracy": 0.19346691966056823,
|
|
"num_tokens": 1483587.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.924755525588989,
|
|
"epoch": 0.039103086909403735,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 5.6987,
|
|
"mean_token_accuracy": 0.188627889752388,
|
|
"num_tokens": 1488152.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.852688217163086,
|
|
"epoch": 0.03921481001485917,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 5.7215,
|
|
"mean_token_accuracy": 0.18358622789382933,
|
|
"num_tokens": 1492511.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.978724813461303,
|
|
"epoch": 0.03932653312031461,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.18464895635843276,
|
|
"num_tokens": 1496634.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.969769763946533,
|
|
"epoch": 0.03943825622577005,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 5.7043,
|
|
"mean_token_accuracy": 0.18896432220935822,
|
|
"num_tokens": 1500904.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.869477558135986,
|
|
"epoch": 0.03954997933122549,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 5.618,
|
|
"mean_token_accuracy": 0.19045006930828096,
|
|
"num_tokens": 1504612.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.796671152114868,
|
|
"epoch": 0.03966170243668093,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 5.6544,
|
|
"mean_token_accuracy": 0.1906888410449028,
|
|
"num_tokens": 1508867.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.729340410232544,
|
|
"epoch": 0.03977342554213637,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 5.4135,
|
|
"mean_token_accuracy": 0.21768646389245988,
|
|
"num_tokens": 1512355.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.7904619693756105,
|
|
"epoch": 0.03988514864759181,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 5.6296,
|
|
"mean_token_accuracy": 0.19588664174079895,
|
|
"num_tokens": 1516609.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.907398986816406,
|
|
"epoch": 0.039996871753047246,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 5.7548,
|
|
"mean_token_accuracy": 0.18738234639167786,
|
|
"num_tokens": 1520945.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.954821968078614,
|
|
"epoch": 0.04010859485850269,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 5.7919,
|
|
"mean_token_accuracy": 0.17330280616879462,
|
|
"num_tokens": 1525136.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.911829948425293,
|
|
"epoch": 0.04022031796395813,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 5.7335,
|
|
"mean_token_accuracy": 0.17516723573207854,
|
|
"num_tokens": 1529088.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.8669532299041744,
|
|
"epoch": 0.040332041069413564,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 5.7184,
|
|
"mean_token_accuracy": 0.1801592469215393,
|
|
"num_tokens": 1533923.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.9007471084594725,
|
|
"epoch": 0.040443764174869,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 5.6071,
|
|
"mean_token_accuracy": 0.2007177233695984,
|
|
"num_tokens": 1538031.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.870078706741333,
|
|
"epoch": 0.040555487280324445,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 5.7082,
|
|
"mean_token_accuracy": 0.17969376146793364,
|
|
"num_tokens": 1542217.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.898741054534912,
|
|
"epoch": 0.04066721038577988,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 5.6301,
|
|
"mean_token_accuracy": 0.1878265455365181,
|
|
"num_tokens": 1546102.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.808369493484497,
|
|
"epoch": 0.04077893349123532,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 5.5982,
|
|
"mean_token_accuracy": 0.19695605784654618,
|
|
"num_tokens": 1550292.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.843883419036866,
|
|
"epoch": 0.040890656596690764,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 5.6667,
|
|
"mean_token_accuracy": 0.18189924210309982,
|
|
"num_tokens": 1554231.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.756465053558349,
|
|
"epoch": 0.0410023797021462,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 5.6449,
|
|
"mean_token_accuracy": 0.19178105145692825,
|
|
"num_tokens": 1558101.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.7194788455963135,
|
|
"epoch": 0.04111410280760164,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 5.5983,
|
|
"mean_token_accuracy": 0.19210630059242248,
|
|
"num_tokens": 1562735.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.858875274658203,
|
|
"epoch": 0.04122582591305708,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 5.5543,
|
|
"mean_token_accuracy": 0.2053885355591774,
|
|
"num_tokens": 1566953.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.8189124584198,
|
|
"epoch": 0.04133754901851252,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.6008,
|
|
"mean_token_accuracy": 0.1976646065711975,
|
|
"num_tokens": 1571389.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.766389560699463,
|
|
"epoch": 0.041449272123967956,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 5.7083,
|
|
"mean_token_accuracy": 0.18083323091268538,
|
|
"num_tokens": 1575737.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.973462867736816,
|
|
"epoch": 0.0415609952294234,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 5.7318,
|
|
"mean_token_accuracy": 0.18451347351074218,
|
|
"num_tokens": 1580371.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.9336131572723385,
|
|
"epoch": 0.04167271833487884,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 5.7415,
|
|
"mean_token_accuracy": 0.1852435752749443,
|
|
"num_tokens": 1584569.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.808266925811767,
|
|
"epoch": 0.041784441440334275,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.6746,
|
|
"mean_token_accuracy": 0.1897304803133011,
|
|
"num_tokens": 1588869.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.92993016242981,
|
|
"epoch": 0.04189616454578971,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 5.699,
|
|
"mean_token_accuracy": 0.18758964240550996,
|
|
"num_tokens": 1593254.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.851748323440551,
|
|
"epoch": 0.042007887651245156,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 5.5884,
|
|
"mean_token_accuracy": 0.1970451608300209,
|
|
"num_tokens": 1597506.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.822835302352905,
|
|
"epoch": 0.04211961075670059,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.18840789794921875,
|
|
"num_tokens": 1601930.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.930711030960083,
|
|
"epoch": 0.04223133386215603,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 5.7416,
|
|
"mean_token_accuracy": 0.1909443974494934,
|
|
"num_tokens": 1606634.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.928756856918335,
|
|
"epoch": 0.042343056967611474,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 5.636,
|
|
"mean_token_accuracy": 0.1966735064983368,
|
|
"num_tokens": 1610457.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.8650257110595705,
|
|
"epoch": 0.04245478007306691,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.19099107682704924,
|
|
"num_tokens": 1614315.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.836633396148682,
|
|
"epoch": 0.04256650317852235,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.6984,
|
|
"mean_token_accuracy": 0.1857185408473015,
|
|
"num_tokens": 1618613.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.989507627487183,
|
|
"epoch": 0.04267822628397779,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 5.5825,
|
|
"mean_token_accuracy": 0.19654418975114823,
|
|
"num_tokens": 1622874.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.865914678573608,
|
|
"epoch": 0.04278994938943323,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.5832,
|
|
"mean_token_accuracy": 0.19483028203248978,
|
|
"num_tokens": 1626799.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.614121818542481,
|
|
"epoch": 0.04290167249488867,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 5.6461,
|
|
"mean_token_accuracy": 0.1878322646021843,
|
|
"num_tokens": 1631082.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.902209091186523,
|
|
"epoch": 0.043013395600344104,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 5.6043,
|
|
"mean_token_accuracy": 0.194015434384346,
|
|
"num_tokens": 1635075.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.905219459533692,
|
|
"epoch": 0.04312511870579955,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 5.655,
|
|
"mean_token_accuracy": 0.18685375303030013,
|
|
"num_tokens": 1639663.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.824585723876953,
|
|
"epoch": 0.043236841811254985,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.7887,
|
|
"mean_token_accuracy": 0.17511658370494843,
|
|
"num_tokens": 1644285.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.836113452911377,
|
|
"epoch": 0.04334856491671042,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 5.6278,
|
|
"mean_token_accuracy": 0.19246720820665358,
|
|
"num_tokens": 1648542.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.852768421173096,
|
|
"epoch": 0.043460288022165866,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 5.688,
|
|
"mean_token_accuracy": 0.18379789143800734,
|
|
"num_tokens": 1652982.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.867275238037109,
|
|
"epoch": 0.0435720111276213,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.5677,
|
|
"mean_token_accuracy": 0.19981538355350495,
|
|
"num_tokens": 1657111.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.74615364074707,
|
|
"epoch": 0.04368373423307674,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.5114,
|
|
"mean_token_accuracy": 0.20943004339933396,
|
|
"num_tokens": 1661541.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.7881919860839846,
|
|
"epoch": 0.043795457338532184,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 5.6962,
|
|
"mean_token_accuracy": 0.1931190684437752,
|
|
"num_tokens": 1665606.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.764306926727295,
|
|
"epoch": 0.04390718044398762,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.5577,
|
|
"mean_token_accuracy": 0.20224445462226867,
|
|
"num_tokens": 1670110.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.783484220504761,
|
|
"epoch": 0.04401890354944306,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.6797,
|
|
"mean_token_accuracy": 0.19021057486534118,
|
|
"num_tokens": 1674724.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.712728071212768,
|
|
"epoch": 0.0441306266548985,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.4867,
|
|
"mean_token_accuracy": 0.20342362821102142,
|
|
"num_tokens": 1679022.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.82525110244751,
|
|
"epoch": 0.04424234976035394,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.1854166254401207,
|
|
"num_tokens": 1683192.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.9117003917694095,
|
|
"epoch": 0.04435407286580938,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.5508,
|
|
"mean_token_accuracy": 0.1991869479417801,
|
|
"num_tokens": 1687442.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.892691707611084,
|
|
"epoch": 0.044465795971264814,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.6187,
|
|
"mean_token_accuracy": 0.19358569532632827,
|
|
"num_tokens": 1691528.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.792757225036621,
|
|
"epoch": 0.04457751907672026,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 5.7013,
|
|
"mean_token_accuracy": 0.18120188936591147,
|
|
"num_tokens": 1695936.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.806407785415649,
|
|
"epoch": 0.044689242182175695,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.647,
|
|
"mean_token_accuracy": 0.1883596122264862,
|
|
"num_tokens": 1700729.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.762160444259644,
|
|
"epoch": 0.04480096528763113,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 5.5854,
|
|
"mean_token_accuracy": 0.1900200754404068,
|
|
"num_tokens": 1704962.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.903286409378052,
|
|
"epoch": 0.044912688393086576,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 5.7253,
|
|
"mean_token_accuracy": 0.188412207365036,
|
|
"num_tokens": 1709408.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.749392414093018,
|
|
"epoch": 0.045024411498542014,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 5.544,
|
|
"mean_token_accuracy": 0.20539478808641434,
|
|
"num_tokens": 1713894.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.758798837661743,
|
|
"epoch": 0.04513613460399745,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 5.4963,
|
|
"mean_token_accuracy": 0.19964819103479386,
|
|
"num_tokens": 1718477.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.695912027359009,
|
|
"epoch": 0.045247857709452895,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 5.4753,
|
|
"mean_token_accuracy": 0.2068771407008171,
|
|
"num_tokens": 1722438.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.75974669456482,
|
|
"epoch": 0.04535958081490833,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 5.5831,
|
|
"mean_token_accuracy": 0.19000767022371293,
|
|
"num_tokens": 1726360.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.711872291564942,
|
|
"epoch": 0.04547130392036377,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 5.5884,
|
|
"mean_token_accuracy": 0.19782641232013704,
|
|
"num_tokens": 1730550.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.848684644699096,
|
|
"epoch": 0.04558302702581921,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 5.6181,
|
|
"mean_token_accuracy": 0.19873436838388442,
|
|
"num_tokens": 1734685.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.871687984466552,
|
|
"epoch": 0.04569475013127465,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.6398,
|
|
"mean_token_accuracy": 0.18782007247209548,
|
|
"num_tokens": 1738714.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.738994503021241,
|
|
"epoch": 0.04580647323673009,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.19532090276479722,
|
|
"num_tokens": 1742714.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.682830095291138,
|
|
"epoch": 0.045918196342185524,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 5.5304,
|
|
"mean_token_accuracy": 0.19776215255260468,
|
|
"num_tokens": 1747053.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.786202430725098,
|
|
"epoch": 0.04602991944764097,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 5.6494,
|
|
"mean_token_accuracy": 0.18340060263872146,
|
|
"num_tokens": 1750766.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.702880001068115,
|
|
"epoch": 0.046141642553096406,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 5.5526,
|
|
"mean_token_accuracy": 0.20375038385391236,
|
|
"num_tokens": 1754835.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.871013212203979,
|
|
"epoch": 0.04625336565855184,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 5.6323,
|
|
"mean_token_accuracy": 0.19306133091449737,
|
|
"num_tokens": 1759088.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.733165454864502,
|
|
"epoch": 0.04636508876400729,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 5.4566,
|
|
"mean_token_accuracy": 0.19918377846479415,
|
|
"num_tokens": 1763493.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.687415218353271,
|
|
"epoch": 0.046476811869462724,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 5.6178,
|
|
"mean_token_accuracy": 0.19442993849515916,
|
|
"num_tokens": 1767859.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.834854173660278,
|
|
"epoch": 0.04658853497491816,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 5.6958,
|
|
"mean_token_accuracy": 0.1966529980301857,
|
|
"num_tokens": 1772448.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.876561212539673,
|
|
"epoch": 0.046700258080373605,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.19237288534641267,
|
|
"num_tokens": 1776420.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.78769850730896,
|
|
"epoch": 0.04681198118582904,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.20187776684761047,
|
|
"num_tokens": 1780503.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.806005382537842,
|
|
"epoch": 0.04692370429128448,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 5.5254,
|
|
"mean_token_accuracy": 0.18709133192896843,
|
|
"num_tokens": 1784730.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.821065235137939,
|
|
"epoch": 0.047035427396739916,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 5.5357,
|
|
"mean_token_accuracy": 0.19565414786338806,
|
|
"num_tokens": 1789317.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.8124915599823,
|
|
"epoch": 0.04714715050219536,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 5.7168,
|
|
"mean_token_accuracy": 0.18089606761932372,
|
|
"num_tokens": 1793431.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.83145956993103,
|
|
"epoch": 0.0472588736076508,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 5.5886,
|
|
"mean_token_accuracy": 0.19206707179546356,
|
|
"num_tokens": 1797546.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.815211296081543,
|
|
"epoch": 0.047370596713106235,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.19378799498081206,
|
|
"num_tokens": 1801995.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.810037565231323,
|
|
"epoch": 0.04748231981856168,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 5.5494,
|
|
"mean_token_accuracy": 0.1948327451944351,
|
|
"num_tokens": 1805957.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.692611360549927,
|
|
"epoch": 0.047594042924017116,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.20250043869018555,
|
|
"num_tokens": 1810216.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.737909555435181,
|
|
"epoch": 0.04770576602947255,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 5.5878,
|
|
"mean_token_accuracy": 0.19820891618728637,
|
|
"num_tokens": 1814727.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.843041563034058,
|
|
"epoch": 0.047817489134928,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.2012895479798317,
|
|
"num_tokens": 1819076.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.887670660018921,
|
|
"epoch": 0.047929212240383434,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 5.6315,
|
|
"mean_token_accuracy": 0.19086363166570663,
|
|
"num_tokens": 1823289.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.619873428344727,
|
|
"epoch": 0.04804093534583887,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 5.4709,
|
|
"mean_token_accuracy": 0.2047662228345871,
|
|
"num_tokens": 1827301.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.79863338470459,
|
|
"epoch": 0.048152658451294315,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 5.6093,
|
|
"mean_token_accuracy": 0.19400935769081115,
|
|
"num_tokens": 1831300.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.751356601715088,
|
|
"epoch": 0.04826438155674975,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 5.4207,
|
|
"mean_token_accuracy": 0.20203897207975388,
|
|
"num_tokens": 1835445.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.778922367095947,
|
|
"epoch": 0.04837610466220519,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 5.5592,
|
|
"mean_token_accuracy": 0.19020436555147172,
|
|
"num_tokens": 1839773.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.876074647903442,
|
|
"epoch": 0.04848782776766063,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.19834407269954682,
|
|
"num_tokens": 1844011.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.866834402084351,
|
|
"epoch": 0.04859955087311607,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 5.6563,
|
|
"mean_token_accuracy": 0.18741424530744552,
|
|
"num_tokens": 1848063.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.783164978027344,
|
|
"epoch": 0.04871127397857151,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.18896054923534394,
|
|
"num_tokens": 1852403.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.707462406158447,
|
|
"epoch": 0.048822997084026945,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.5782,
|
|
"mean_token_accuracy": 0.19232478588819504,
|
|
"num_tokens": 1856680.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.941496753692627,
|
|
"epoch": 0.04893472018948239,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 5.6988,
|
|
"mean_token_accuracy": 0.18841493278741836,
|
|
"num_tokens": 1861166.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.853933238983155,
|
|
"epoch": 0.049046443294937826,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 5.5343,
|
|
"mean_token_accuracy": 0.19651836454868316,
|
|
"num_tokens": 1865334.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.7885599613189695,
|
|
"epoch": 0.04915816640039326,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 5.5103,
|
|
"mean_token_accuracy": 0.20324531346559524,
|
|
"num_tokens": 1869391.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.779803848266601,
|
|
"epoch": 0.04926988950584871,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.18151400834321976,
|
|
"num_tokens": 1873811.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.938524341583252,
|
|
"epoch": 0.049381612611304145,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 5.6113,
|
|
"mean_token_accuracy": 0.197547022998333,
|
|
"num_tokens": 1878269.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.9333361148834225,
|
|
"epoch": 0.04949333571675958,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 5.72,
|
|
"mean_token_accuracy": 0.1759050115942955,
|
|
"num_tokens": 1882800.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.799074602127075,
|
|
"epoch": 0.049605058822215026,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 5.6057,
|
|
"mean_token_accuracy": 0.19079000651836395,
|
|
"num_tokens": 1886968.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.835526609420777,
|
|
"epoch": 0.04971678192767046,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 5.6423,
|
|
"mean_token_accuracy": 0.18531539142131806,
|
|
"num_tokens": 1891158.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.666534328460694,
|
|
"epoch": 0.0498285050331259,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 5.3402,
|
|
"mean_token_accuracy": 0.2212604433298111,
|
|
"num_tokens": 1895431.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.787649869918823,
|
|
"epoch": 0.04994022813858134,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 5.5506,
|
|
"mean_token_accuracy": 0.1960133582353592,
|
|
"num_tokens": 1899750.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.787043285369873,
|
|
"epoch": 0.05005195124403678,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 5.4654,
|
|
"mean_token_accuracy": 0.20220616459846497,
|
|
"num_tokens": 1904076.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.723090791702271,
|
|
"epoch": 0.05016367434949222,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.5961,
|
|
"mean_token_accuracy": 0.1981295756995678,
|
|
"num_tokens": 1908859.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.7092540740966795,
|
|
"epoch": 0.050275397454947655,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 5.3811,
|
|
"mean_token_accuracy": 0.21009975224733352,
|
|
"num_tokens": 1913049.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.761197090148926,
|
|
"epoch": 0.0503871205604031,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 5.5858,
|
|
"mean_token_accuracy": 0.1940508484840393,
|
|
"num_tokens": 1917247.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.7653045654296875,
|
|
"epoch": 0.05049884366585854,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 5.5715,
|
|
"mean_token_accuracy": 0.2005481407046318,
|
|
"num_tokens": 1920958.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.650889492034912,
|
|
"epoch": 0.050610566771313974,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.19745519757270813,
|
|
"num_tokens": 1925184.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.788166809082031,
|
|
"epoch": 0.05072228987676942,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.20047650933265687,
|
|
"num_tokens": 1929392.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.705606317520141,
|
|
"epoch": 0.050834012982224855,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 5.5656,
|
|
"mean_token_accuracy": 0.2057395651936531,
|
|
"num_tokens": 1933535.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.746721363067627,
|
|
"epoch": 0.05094573608768029,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.1967824012041092,
|
|
"num_tokens": 1937830.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.682202482223511,
|
|
"epoch": 0.05105745919313573,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 5.469,
|
|
"mean_token_accuracy": 0.20526605397462844,
|
|
"num_tokens": 1941521.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.666811180114746,
|
|
"epoch": 0.05116918229859117,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 5.5247,
|
|
"mean_token_accuracy": 0.19909686148166655,
|
|
"num_tokens": 1945662.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.675921297073364,
|
|
"epoch": 0.05128090540404661,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 5.3343,
|
|
"mean_token_accuracy": 0.20175234079360962,
|
|
"num_tokens": 1949529.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.5916831493377686,
|
|
"epoch": 0.05139262850950205,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 5.3595,
|
|
"mean_token_accuracy": 0.21346658766269683,
|
|
"num_tokens": 1953541.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.603648614883423,
|
|
"epoch": 0.05150435161495749,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 5.429,
|
|
"mean_token_accuracy": 0.20238481909036637,
|
|
"num_tokens": 1957863.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.6920647621154785,
|
|
"epoch": 0.05161607472041293,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 5.4622,
|
|
"mean_token_accuracy": 0.20448255985975267,
|
|
"num_tokens": 1962085.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.818530702590943,
|
|
"epoch": 0.051727797825868366,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 5.46,
|
|
"mean_token_accuracy": 0.20918918401002884,
|
|
"num_tokens": 1966707.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.782559156417847,
|
|
"epoch": 0.05183952093132381,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 5.5782,
|
|
"mean_token_accuracy": 0.18843235522508622,
|
|
"num_tokens": 1970461.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.730897331237793,
|
|
"epoch": 0.05195124403677925,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.19218226224184037,
|
|
"num_tokens": 1974611.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.58878173828125,
|
|
"epoch": 0.052062967142234684,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 5.3545,
|
|
"mean_token_accuracy": 0.21814181208610534,
|
|
"num_tokens": 1978644.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.569992017745972,
|
|
"epoch": 0.05217469024769013,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 5.3576,
|
|
"mean_token_accuracy": 0.2130546897649765,
|
|
"num_tokens": 1983045.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.643571615219116,
|
|
"epoch": 0.052286413353145565,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 5.4264,
|
|
"mean_token_accuracy": 0.19753186404705048,
|
|
"num_tokens": 1987046.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.820702457427979,
|
|
"epoch": 0.052398136458601,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 5.6737,
|
|
"mean_token_accuracy": 0.19394189715385438,
|
|
"num_tokens": 1992201.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.794189882278443,
|
|
"epoch": 0.05250985956405644,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 5.5028,
|
|
"mean_token_accuracy": 0.20308198630809784,
|
|
"num_tokens": 1996124.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.6884153366088865,
|
|
"epoch": 0.052621582669511884,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 5.5133,
|
|
"mean_token_accuracy": 0.19611742794513704,
|
|
"num_tokens": 2000602.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.69493989944458,
|
|
"epoch": 0.05273330577496732,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 5.5597,
|
|
"mean_token_accuracy": 0.1906090185046196,
|
|
"num_tokens": 2005538.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.730038738250732,
|
|
"epoch": 0.05284502888042276,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 5.5006,
|
|
"mean_token_accuracy": 0.19747251272201538,
|
|
"num_tokens": 2010158.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.713155364990234,
|
|
"epoch": 0.0529567519858782,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 5.4504,
|
|
"mean_token_accuracy": 0.20709324926137923,
|
|
"num_tokens": 2013993.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.664393043518066,
|
|
"epoch": 0.05306847509133364,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.20383214950561523,
|
|
"num_tokens": 2018312.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.771481895446778,
|
|
"epoch": 0.053180198196789076,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 5.6226,
|
|
"mean_token_accuracy": 0.19073883444070816,
|
|
"num_tokens": 2022957.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.73010287284851,
|
|
"epoch": 0.05329192130224452,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 5.548,
|
|
"mean_token_accuracy": 0.19659071415662766,
|
|
"num_tokens": 2027546.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.743465614318848,
|
|
"epoch": 0.05340364440769996,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 5.362,
|
|
"mean_token_accuracy": 0.19935485869646072,
|
|
"num_tokens": 2031370.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.73956298828125,
|
|
"epoch": 0.053515367513155394,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 5.5286,
|
|
"mean_token_accuracy": 0.19233475774526596,
|
|
"num_tokens": 2036314.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.790511226654052,
|
|
"epoch": 0.05362709061861083,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 5.4899,
|
|
"mean_token_accuracy": 0.1918132334947586,
|
|
"num_tokens": 2040819.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.736824607849121,
|
|
"epoch": 0.053738813724066276,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 5.4628,
|
|
"mean_token_accuracy": 0.20366894453763962,
|
|
"num_tokens": 2045230.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.79299521446228,
|
|
"epoch": 0.05385053682952171,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 5.4941,
|
|
"mean_token_accuracy": 0.20016811043024063,
|
|
"num_tokens": 2049688.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.674882650375366,
|
|
"epoch": 0.05396225993497715,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 5.4647,
|
|
"mean_token_accuracy": 0.20398021638393402,
|
|
"num_tokens": 2053811.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.634827518463135,
|
|
"epoch": 0.054073983040432594,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 5.5145,
|
|
"mean_token_accuracy": 0.19332455545663835,
|
|
"num_tokens": 2058018.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.709335565567017,
|
|
"epoch": 0.05418570614588803,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 5.5103,
|
|
"mean_token_accuracy": 0.1937290146946907,
|
|
"num_tokens": 2062258.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.741044902801514,
|
|
"epoch": 0.05429742925134347,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 5.5401,
|
|
"mean_token_accuracy": 0.1956827312707901,
|
|
"num_tokens": 2066465.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.652325916290283,
|
|
"epoch": 0.05440915235679891,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 5.3751,
|
|
"mean_token_accuracy": 0.21417505145072938,
|
|
"num_tokens": 2070679.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.665175867080689,
|
|
"epoch": 0.05452087546225435,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 5.4719,
|
|
"mean_token_accuracy": 0.19800220280885697,
|
|
"num_tokens": 2075443.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.677209901809692,
|
|
"epoch": 0.054632598567709786,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 5.4908,
|
|
"mean_token_accuracy": 0.18837572783231735,
|
|
"num_tokens": 2079726.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.596740770339966,
|
|
"epoch": 0.05474432167316523,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.4139,
|
|
"mean_token_accuracy": 0.20456188321113586,
|
|
"num_tokens": 2083868.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.6920839786529545,
|
|
"epoch": 0.05485604477862067,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 5.4324,
|
|
"mean_token_accuracy": 0.20738107711076736,
|
|
"num_tokens": 2087722.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.734287595748901,
|
|
"epoch": 0.054967767884076105,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.3726,
|
|
"mean_token_accuracy": 0.21619302332401275,
|
|
"num_tokens": 2092293.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.639990568161011,
|
|
"epoch": 0.05507949098953154,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 5.4867,
|
|
"mean_token_accuracy": 0.21042072176933288,
|
|
"num_tokens": 2096715.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.737297534942627,
|
|
"epoch": 0.055191214094986986,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 5.5661,
|
|
"mean_token_accuracy": 0.19853804111480713,
|
|
"num_tokens": 2100810.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.691304540634155,
|
|
"epoch": 0.05530293720044242,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 5.4729,
|
|
"mean_token_accuracy": 0.19330257475376128,
|
|
"num_tokens": 2105125.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.716330480575562,
|
|
"epoch": 0.05541466030589786,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 5.5589,
|
|
"mean_token_accuracy": 0.19316209107637405,
|
|
"num_tokens": 2109218.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.671482372283935,
|
|
"epoch": 0.055526383411353304,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 5.4623,
|
|
"mean_token_accuracy": 0.19808712005615234,
|
|
"num_tokens": 2113361.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.645354509353638,
|
|
"epoch": 0.05563810651680874,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 5.4073,
|
|
"mean_token_accuracy": 0.20510194897651673,
|
|
"num_tokens": 2117457.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.7331404209136965,
|
|
"epoch": 0.05574982962226418,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 5.4521,
|
|
"mean_token_accuracy": 0.21233909726142883,
|
|
"num_tokens": 2121198.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.696898126602173,
|
|
"epoch": 0.05586155272771962,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 5.5453,
|
|
"mean_token_accuracy": 0.19986631721258163,
|
|
"num_tokens": 2125742.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.617610836029053,
|
|
"epoch": 0.05597327583317506,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 5.4545,
|
|
"mean_token_accuracy": 0.1947040691971779,
|
|
"num_tokens": 2129851.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.578350973129273,
|
|
"epoch": 0.0560849989386305,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 5.241,
|
|
"mean_token_accuracy": 0.2216262102127075,
|
|
"num_tokens": 2133814.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.724449157714844,
|
|
"epoch": 0.05619672204408594,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 5.5208,
|
|
"mean_token_accuracy": 0.20132671147584916,
|
|
"num_tokens": 2138191.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.729740953445434,
|
|
"epoch": 0.05630844514954138,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.19500400125980377,
|
|
"num_tokens": 2142144.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.7269703388214115,
|
|
"epoch": 0.056420168254996815,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 5.4114,
|
|
"mean_token_accuracy": 0.1983845517039299,
|
|
"num_tokens": 2146308.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.745017242431641,
|
|
"epoch": 0.05653189136045225,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 5.5356,
|
|
"mean_token_accuracy": 0.187234328687191,
|
|
"num_tokens": 2150810.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.739377689361572,
|
|
"epoch": 0.056643614465907696,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.18481508195400237,
|
|
"num_tokens": 2154865.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.647379541397095,
|
|
"epoch": 0.05675533757136313,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 5.3846,
|
|
"mean_token_accuracy": 0.2152353584766388,
|
|
"num_tokens": 2159045.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.694047212600708,
|
|
"epoch": 0.05686706067681857,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 5.43,
|
|
"mean_token_accuracy": 0.20127338021993638,
|
|
"num_tokens": 2163151.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.623006582260132,
|
|
"epoch": 0.056978783782274015,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 5.4784,
|
|
"mean_token_accuracy": 0.20224980711936952,
|
|
"num_tokens": 2167099.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.720960998535157,
|
|
"epoch": 0.05709050688772945,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 5.5428,
|
|
"mean_token_accuracy": 0.19093946367502213,
|
|
"num_tokens": 2171299.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.754732465744018,
|
|
"epoch": 0.05720222999318489,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 5.5369,
|
|
"mean_token_accuracy": 0.19886625409126282,
|
|
"num_tokens": 2175491.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.778254556655884,
|
|
"epoch": 0.05731395309864033,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 5.3294,
|
|
"mean_token_accuracy": 0.19855942130088805,
|
|
"num_tokens": 2179677.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.752972841262817,
|
|
"epoch": 0.05742567620409577,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 5.4079,
|
|
"mean_token_accuracy": 0.20372159034013748,
|
|
"num_tokens": 2184038.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.733305263519287,
|
|
"epoch": 0.05753739930955121,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.19013774991035462,
|
|
"num_tokens": 2188278.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.702356815338135,
|
|
"epoch": 0.057649122415006644,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 5.6095,
|
|
"mean_token_accuracy": 0.19494371265172958,
|
|
"num_tokens": 2192402.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.578436946868896,
|
|
"epoch": 0.05776084552046209,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 5.3235,
|
|
"mean_token_accuracy": 0.21250910609960555,
|
|
"num_tokens": 2196631.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.638101482391358,
|
|
"epoch": 0.057872568625917525,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.20848297327756882,
|
|
"num_tokens": 2201016.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.593278789520264,
|
|
"epoch": 0.05798429173137296,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 5.3912,
|
|
"mean_token_accuracy": 0.21250712126493454,
|
|
"num_tokens": 2205091.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.684863185882568,
|
|
"epoch": 0.05809601483682841,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.20237696319818496,
|
|
"num_tokens": 2209525.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.655977392196656,
|
|
"epoch": 0.058207737942283844,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 5.409,
|
|
"mean_token_accuracy": 0.20060928910970688,
|
|
"num_tokens": 2214093.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.663346529006958,
|
|
"epoch": 0.05831946104773928,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.19938003718852998,
|
|
"num_tokens": 2218619.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.671671676635742,
|
|
"epoch": 0.058431184153194725,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 5.3489,
|
|
"mean_token_accuracy": 0.20987917482852936,
|
|
"num_tokens": 2222953.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.648585557937622,
|
|
"epoch": 0.05854290725865016,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 5.4249,
|
|
"mean_token_accuracy": 0.2062181681394577,
|
|
"num_tokens": 2227162.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.658475542068482,
|
|
"epoch": 0.0586546303641056,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 5.3822,
|
|
"mean_token_accuracy": 0.19644873440265656,
|
|
"num_tokens": 2231333.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.695336866378784,
|
|
"epoch": 0.05876635346956104,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.19799769371747972,
|
|
"num_tokens": 2235160.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.700191783905029,
|
|
"epoch": 0.05887807657501648,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 5.4899,
|
|
"mean_token_accuracy": 0.20430560261011124,
|
|
"num_tokens": 2239699.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.764710092544556,
|
|
"epoch": 0.05898979968047192,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 5.4771,
|
|
"mean_token_accuracy": 0.1954497367143631,
|
|
"num_tokens": 2244146.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.732822465896606,
|
|
"epoch": 0.059101522785927355,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.19945912808179855,
|
|
"num_tokens": 2248190.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.663884496688842,
|
|
"epoch": 0.0592132458913828,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 5.3814,
|
|
"mean_token_accuracy": 0.2188004434108734,
|
|
"num_tokens": 2252389.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.584512186050415,
|
|
"epoch": 0.059324968996838236,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 5.3835,
|
|
"mean_token_accuracy": 0.21272893846035004,
|
|
"num_tokens": 2256430.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.609366416931152,
|
|
"epoch": 0.05943669210229367,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.20157335847616195,
|
|
"num_tokens": 2260632.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.64447250366211,
|
|
"epoch": 0.05954841520774912,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 5.374,
|
|
"mean_token_accuracy": 0.20789485424757004,
|
|
"num_tokens": 2264593.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.657564306259156,
|
|
"epoch": 0.059660138313204554,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 5.3631,
|
|
"mean_token_accuracy": 0.20994101464748383,
|
|
"num_tokens": 2268994.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.724624824523926,
|
|
"epoch": 0.05977186141865999,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 5.4608,
|
|
"mean_token_accuracy": 0.20652345567941666,
|
|
"num_tokens": 2273355.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.622627449035645,
|
|
"epoch": 0.059883584524115435,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 5.4408,
|
|
"mean_token_accuracy": 0.21060604006052017,
|
|
"num_tokens": 2277618.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.578470468521118,
|
|
"epoch": 0.05999530762957087,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.20465112030506133,
|
|
"num_tokens": 2281751.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.6411576747894285,
|
|
"epoch": 0.06010703073502631,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.19300513565540314,
|
|
"num_tokens": 2286581.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.68374400138855,
|
|
"epoch": 0.06021875384048175,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 5.4147,
|
|
"mean_token_accuracy": 0.20215319246053695,
|
|
"num_tokens": 2290733.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.723708868026733,
|
|
"epoch": 0.06033047694593719,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 5.3536,
|
|
"mean_token_accuracy": 0.20597321391105652,
|
|
"num_tokens": 2295150.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.7238153457641605,
|
|
"epoch": 0.06044220005139263,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 5.4375,
|
|
"mean_token_accuracy": 0.20747863054275512,
|
|
"num_tokens": 2299461.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.587930822372437,
|
|
"epoch": 0.060553923156848065,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 5.2821,
|
|
"mean_token_accuracy": 0.22456400692462922,
|
|
"num_tokens": 2303981.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.83951325416565,
|
|
"epoch": 0.06066564626230351,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 5.6461,
|
|
"mean_token_accuracy": 0.18310097455978394,
|
|
"num_tokens": 2308514.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.807481193542481,
|
|
"epoch": 0.060777369367758946,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 5.5157,
|
|
"mean_token_accuracy": 0.20032941550016403,
|
|
"num_tokens": 2312803.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.706876802444458,
|
|
"epoch": 0.06088909247321438,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 5.4204,
|
|
"mean_token_accuracy": 0.20374609380960465,
|
|
"num_tokens": 2317136.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.711637687683106,
|
|
"epoch": 0.06100081557866983,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 5.4106,
|
|
"mean_token_accuracy": 0.20451716631650924,
|
|
"num_tokens": 2321064.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.683818435668945,
|
|
"epoch": 0.061112538684125264,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 5.3412,
|
|
"mean_token_accuracy": 0.21318926066160201,
|
|
"num_tokens": 2324971.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.679719495773315,
|
|
"epoch": 0.0612242617895807,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 5.4779,
|
|
"mean_token_accuracy": 0.19736759066581727,
|
|
"num_tokens": 2329433.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.600196123123169,
|
|
"epoch": 0.061335984895036146,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 5.3563,
|
|
"mean_token_accuracy": 0.2150166630744934,
|
|
"num_tokens": 2333493.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.57360200881958,
|
|
"epoch": 0.06144770800049158,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 5.2304,
|
|
"mean_token_accuracy": 0.21621160507202147,
|
|
"num_tokens": 2337663.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.678966856002807,
|
|
"epoch": 0.06155943110594702,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 5.2662,
|
|
"mean_token_accuracy": 0.2184838131070137,
|
|
"num_tokens": 2342291.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.676230812072754,
|
|
"epoch": 0.06167115421140246,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 5.4342,
|
|
"mean_token_accuracy": 0.20801472216844558,
|
|
"num_tokens": 2346660.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.669618463516235,
|
|
"epoch": 0.0617828773168579,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 5.3922,
|
|
"mean_token_accuracy": 0.20119996964931489,
|
|
"num_tokens": 2350605.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.683516979217529,
|
|
"epoch": 0.06189460042231334,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 5.3747,
|
|
"mean_token_accuracy": 0.20869618207216262,
|
|
"num_tokens": 2355125.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.705159473419189,
|
|
"epoch": 0.062006323527768775,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.194733664393425,
|
|
"num_tokens": 2359058.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.627893495559692,
|
|
"epoch": 0.06211804663322422,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 5.463,
|
|
"mean_token_accuracy": 0.21798547804355622,
|
|
"num_tokens": 2363326.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.7318541526794435,
|
|
"epoch": 0.062229769738679656,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 5.2807,
|
|
"mean_token_accuracy": 0.21216789335012437,
|
|
"num_tokens": 2367241.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.78978590965271,
|
|
"epoch": 0.062341492844135094,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 5.4647,
|
|
"mean_token_accuracy": 0.1928023025393486,
|
|
"num_tokens": 2371660.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.7157361030578615,
|
|
"epoch": 0.06245321594959054,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 5.5075,
|
|
"mean_token_accuracy": 0.20195768922567367,
|
|
"num_tokens": 2376083.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.644212436676026,
|
|
"epoch": 0.06256493905504597,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 5.3724,
|
|
"mean_token_accuracy": 0.21047956198453904,
|
|
"num_tokens": 2380006.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.660019779205323,
|
|
"epoch": 0.06267666216050141,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 5.4079,
|
|
"mean_token_accuracy": 0.20587846338748933,
|
|
"num_tokens": 2384382.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.658632707595825,
|
|
"epoch": 0.06278838526595686,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 5.3543,
|
|
"mean_token_accuracy": 0.20837244987487794,
|
|
"num_tokens": 2388550.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.6060553073883055,
|
|
"epoch": 0.06290010837141229,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 5.3939,
|
|
"mean_token_accuracy": 0.20029546022415162,
|
|
"num_tokens": 2392189.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.601018953323364,
|
|
"epoch": 0.06301183147686773,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 5.3059,
|
|
"mean_token_accuracy": 0.22086506187915803,
|
|
"num_tokens": 2396016.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.631944751739502,
|
|
"epoch": 0.06312355458232317,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 5.4003,
|
|
"mean_token_accuracy": 0.20547943860292434,
|
|
"num_tokens": 2400053.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.736320638656617,
|
|
"epoch": 0.0632352776877786,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 5.498,
|
|
"mean_token_accuracy": 0.18274902701377868,
|
|
"num_tokens": 2404164.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.6216331958770756,
|
|
"epoch": 0.06334700079323405,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 5.4779,
|
|
"mean_token_accuracy": 0.2058979496359825,
|
|
"num_tokens": 2408477.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.798714542388916,
|
|
"epoch": 0.06345872389868949,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.19199300259351731,
|
|
"num_tokens": 2413299.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.629748058319092,
|
|
"epoch": 0.06357044700414492,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 5.2433,
|
|
"mean_token_accuracy": 0.2218858554959297,
|
|
"num_tokens": 2417668.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.650796365737915,
|
|
"epoch": 0.06368217010960037,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 5.4082,
|
|
"mean_token_accuracy": 0.20668836534023285,
|
|
"num_tokens": 2422531.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.572155332565307,
|
|
"epoch": 0.06379389321505581,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 5.2035,
|
|
"mean_token_accuracy": 0.222446571290493,
|
|
"num_tokens": 2426958.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.589015960693359,
|
|
"epoch": 0.06390561632051124,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 5.3825,
|
|
"mean_token_accuracy": 0.21515996307134627,
|
|
"num_tokens": 2431011.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.543829345703125,
|
|
"epoch": 0.06401733942596669,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 5.34,
|
|
"mean_token_accuracy": 0.206784251332283,
|
|
"num_tokens": 2435050.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.694529294967651,
|
|
"epoch": 0.06412906253142213,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.2040119305253029,
|
|
"num_tokens": 2440017.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.667089891433716,
|
|
"epoch": 0.06424078563687756,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 5.3921,
|
|
"mean_token_accuracy": 0.20922284126281737,
|
|
"num_tokens": 2444177.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.710749530792237,
|
|
"epoch": 0.064352508742333,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.19369867593050002,
|
|
"num_tokens": 2448287.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.707183218002319,
|
|
"epoch": 0.06446423184778845,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 5.4414,
|
|
"mean_token_accuracy": 0.20463113188743592,
|
|
"num_tokens": 2452901.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.6640900611877445,
|
|
"epoch": 0.06457595495324388,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 5.2873,
|
|
"mean_token_accuracy": 0.22243628948926925,
|
|
"num_tokens": 2457333.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.768609142303466,
|
|
"epoch": 0.06468767805869932,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.19730397388339044,
|
|
"num_tokens": 2461502.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.744687557220459,
|
|
"epoch": 0.06479940116415477,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 5.5625,
|
|
"mean_token_accuracy": 0.19384552538394928,
|
|
"num_tokens": 2465742.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.689142560958862,
|
|
"epoch": 0.0649111242696102,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 5.4767,
|
|
"mean_token_accuracy": 0.19787981808185579,
|
|
"num_tokens": 2469768.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.614039134979248,
|
|
"epoch": 0.06502284737506564,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.2049763947725296,
|
|
"num_tokens": 2473867.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.599695920944214,
|
|
"epoch": 0.06513457048052107,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 5.3444,
|
|
"mean_token_accuracy": 0.20889384001493455,
|
|
"num_tokens": 2477783.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.6680515766143795,
|
|
"epoch": 0.06524629358597651,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 5.5353,
|
|
"mean_token_accuracy": 0.19455234706401825,
|
|
"num_tokens": 2482135.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.678571319580078,
|
|
"epoch": 0.06535801669143196,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 5.466,
|
|
"mean_token_accuracy": 0.19640394151210785,
|
|
"num_tokens": 2486171.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.572101306915283,
|
|
"epoch": 0.06546973979688739,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 5.2695,
|
|
"mean_token_accuracy": 0.21688248813152314,
|
|
"num_tokens": 2490372.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.638118267059326,
|
|
"epoch": 0.06558146290234283,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 5.2066,
|
|
"mean_token_accuracy": 0.2306307077407837,
|
|
"num_tokens": 2494662.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.559669113159179,
|
|
"epoch": 0.06569318600779828,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 5.3592,
|
|
"mean_token_accuracy": 0.21668146252632142,
|
|
"num_tokens": 2498164.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.663913297653198,
|
|
"epoch": 0.0658049091132537,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 5.4256,
|
|
"mean_token_accuracy": 0.1958498999476433,
|
|
"num_tokens": 2502335.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.715854167938232,
|
|
"epoch": 0.06591663221870915,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 5.5713,
|
|
"mean_token_accuracy": 0.1957806333899498,
|
|
"num_tokens": 2506792.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.641438007354736,
|
|
"epoch": 0.0660283553241646,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 5.3534,
|
|
"mean_token_accuracy": 0.21230535209178925,
|
|
"num_tokens": 2510863.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.697113370895385,
|
|
"epoch": 0.06614007842962003,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 5.4717,
|
|
"mean_token_accuracy": 0.19569053649902343,
|
|
"num_tokens": 2515220.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.630364465713501,
|
|
"epoch": 0.06625180153507547,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 5.345,
|
|
"mean_token_accuracy": 0.20241880565881729,
|
|
"num_tokens": 2519501.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.6378789901733395,
|
|
"epoch": 0.06636352464053091,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 5.388,
|
|
"mean_token_accuracy": 0.20747827142477035,
|
|
"num_tokens": 2523838.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.594701528549194,
|
|
"epoch": 0.06647524774598634,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 5.375,
|
|
"mean_token_accuracy": 0.20893704146146774,
|
|
"num_tokens": 2528480.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.649026823043823,
|
|
"epoch": 0.06658697085144179,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 5.4124,
|
|
"mean_token_accuracy": 0.19796885550022125,
|
|
"num_tokens": 2532596.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.646418809890747,
|
|
"epoch": 0.06669869395689723,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 5.3353,
|
|
"mean_token_accuracy": 0.21195341944694518,
|
|
"num_tokens": 2536980.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.716257476806641,
|
|
"epoch": 0.06681041706235266,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 5.4948,
|
|
"mean_token_accuracy": 0.1920865833759308,
|
|
"num_tokens": 2540729.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.747273063659668,
|
|
"epoch": 0.0669221401678081,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 5.5499,
|
|
"mean_token_accuracy": 0.18710134774446488,
|
|
"num_tokens": 2545164.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.672609186172485,
|
|
"epoch": 0.06703386327326355,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 5.2965,
|
|
"mean_token_accuracy": 0.21778741478919983,
|
|
"num_tokens": 2549182.0,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"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": 545606673039360.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|