13178 lines
359 KiB
JSON
13178 lines
359 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.904134484325307,
|
|
"eval_steps": 500,
|
|
"global_step": 6500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.69198923110962,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 14.25,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8255,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10131.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691961860656738,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 13.9375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7543,
|
|
"mean_token_accuracy": 0.00033454534714110195,
|
|
"num_tokens": 20042.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690863990783692,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 9.8125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5076,
|
|
"mean_token_accuracy": 0.02908163322135806,
|
|
"num_tokens": 30596.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.680220794677734,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 6.5625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1604,
|
|
"mean_token_accuracy": 0.04479682110249996,
|
|
"num_tokens": 40667.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.61959457397461,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 4.78125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8647,
|
|
"mean_token_accuracy": 0.04831764362752437,
|
|
"num_tokens": 51027.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.559894180297851,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7234,
|
|
"mean_token_accuracy": 0.043796106800436975,
|
|
"num_tokens": 61860.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.580874252319337,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.5845,
|
|
"mean_token_accuracy": 0.042750259116292,
|
|
"num_tokens": 72136.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.5773193359375,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 2.796875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.561,
|
|
"mean_token_accuracy": 0.0454920094460249,
|
|
"num_tokens": 82611.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.545099830627441,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5763,
|
|
"mean_token_accuracy": 0.04643099419772625,
|
|
"num_tokens": 92722.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.54197654724121,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4838,
|
|
"mean_token_accuracy": 0.051533405855298045,
|
|
"num_tokens": 103258.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.54912166595459,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4581,
|
|
"mean_token_accuracy": 0.050457367300987245,
|
|
"num_tokens": 114277.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.54041051864624,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3354,
|
|
"mean_token_accuracy": 0.06432176940143108,
|
|
"num_tokens": 124098.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.39171543121338,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.2163,
|
|
"mean_token_accuracy": 0.059896310418844224,
|
|
"num_tokens": 133775.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.376233863830567,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.1789,
|
|
"mean_token_accuracy": 0.06541561521589756,
|
|
"num_tokens": 144765.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.435562705993652,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.0237,
|
|
"mean_token_accuracy": 0.07151055298745632,
|
|
"num_tokens": 154641.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.328177547454834,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.9198,
|
|
"mean_token_accuracy": 0.07580552473664284,
|
|
"num_tokens": 164728.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.30659589767456,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 2.796875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.8567,
|
|
"mean_token_accuracy": 0.06854435540735722,
|
|
"num_tokens": 174739.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.201157188415527,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.7127,
|
|
"mean_token_accuracy": 0.07744735479354858,
|
|
"num_tokens": 184580.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.151866340637207,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6842,
|
|
"mean_token_accuracy": 0.07533066719770432,
|
|
"num_tokens": 195243.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.13199291229248,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.5675,
|
|
"mean_token_accuracy": 0.08015098795294762,
|
|
"num_tokens": 205751.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.04837646484375,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.4086,
|
|
"mean_token_accuracy": 0.07969476953148842,
|
|
"num_tokens": 217283.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.001432323455811,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.3112,
|
|
"mean_token_accuracy": 0.07908041924238204,
|
|
"num_tokens": 226909.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.96151065826416,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2717,
|
|
"mean_token_accuracy": 0.0716276228427887,
|
|
"num_tokens": 237343.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.76609468460083,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.0896,
|
|
"mean_token_accuracy": 0.08059490025043488,
|
|
"num_tokens": 246601.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.774349784851074,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.9819,
|
|
"mean_token_accuracy": 0.08241913989186286,
|
|
"num_tokens": 256818.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.546303367614746,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.792,
|
|
"mean_token_accuracy": 0.08451972156763077,
|
|
"num_tokens": 267157.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.346164798736572,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.729,
|
|
"mean_token_accuracy": 0.08054408058524132,
|
|
"num_tokens": 277370.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.197631072998046,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6448,
|
|
"mean_token_accuracy": 0.07991334870457649,
|
|
"num_tokens": 287851.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.952075290679932,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6049,
|
|
"mean_token_accuracy": 0.08717379048466682,
|
|
"num_tokens": 299043.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.753842163085938,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.514,
|
|
"mean_token_accuracy": 0.08479600548744201,
|
|
"num_tokens": 308525.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.543214130401612,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3998,
|
|
"mean_token_accuracy": 0.08677626475691795,
|
|
"num_tokens": 318301.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.366859912872314,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3808,
|
|
"mean_token_accuracy": 0.08776860758662224,
|
|
"num_tokens": 329599.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.208882141113282,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2913,
|
|
"mean_token_accuracy": 0.08820515275001525,
|
|
"num_tokens": 339446.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.189261150360107,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.361,
|
|
"mean_token_accuracy": 0.08676841780543328,
|
|
"num_tokens": 349579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.030198335647583,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.2959,
|
|
"mean_token_accuracy": 0.08807415291666984,
|
|
"num_tokens": 359452.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.909095239639282,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.2007,
|
|
"mean_token_accuracy": 0.09265701398253441,
|
|
"num_tokens": 369558.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.885198545455933,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.2374,
|
|
"mean_token_accuracy": 0.09019400030374528,
|
|
"num_tokens": 380458.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.863682413101197,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2124,
|
|
"mean_token_accuracy": 0.09065315276384353,
|
|
"num_tokens": 390051.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.756366014480591,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.1623,
|
|
"mean_token_accuracy": 0.09404176101088524,
|
|
"num_tokens": 400343.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.77233304977417,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.1858,
|
|
"mean_token_accuracy": 0.09119251668453217,
|
|
"num_tokens": 410004.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.71948881149292,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1628,
|
|
"mean_token_accuracy": 0.08677014708518982,
|
|
"num_tokens": 421286.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.720707845687866,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.1961,
|
|
"mean_token_accuracy": 0.08782461136579514,
|
|
"num_tokens": 432351.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.727673149108886,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1924,
|
|
"mean_token_accuracy": 0.09309446811676025,
|
|
"num_tokens": 442843.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.621369409561157,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.0629,
|
|
"mean_token_accuracy": 0.09052515476942062,
|
|
"num_tokens": 453370.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.63883638381958,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2184,
|
|
"mean_token_accuracy": 0.0872538447380066,
|
|
"num_tokens": 464715.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.644190263748169,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0549,
|
|
"mean_token_accuracy": 0.0908614382147789,
|
|
"num_tokens": 475086.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.491489458084106,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1148,
|
|
"mean_token_accuracy": 0.10133155509829521,
|
|
"num_tokens": 486088.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.578480529785156,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.0797,
|
|
"mean_token_accuracy": 0.09119723215699196,
|
|
"num_tokens": 496978.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.449330139160156,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.864,
|
|
"mean_token_accuracy": 0.1022227019071579,
|
|
"num_tokens": 507794.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.454753732681274,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0454,
|
|
"mean_token_accuracy": 0.09052475318312644,
|
|
"num_tokens": 517763.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.496077680587769,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.9487,
|
|
"mean_token_accuracy": 0.0943218432366848,
|
|
"num_tokens": 527374.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.425622415542603,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.0141,
|
|
"mean_token_accuracy": 0.09675498828291892,
|
|
"num_tokens": 537435.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.358550405502319,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.978,
|
|
"mean_token_accuracy": 0.09722853824496269,
|
|
"num_tokens": 547236.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.3646965503692625,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.954,
|
|
"mean_token_accuracy": 0.09657247290015221,
|
|
"num_tokens": 557130.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.437637186050415,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.968,
|
|
"mean_token_accuracy": 0.09677831679582596,
|
|
"num_tokens": 567482.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.320119571685791,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.862,
|
|
"mean_token_accuracy": 0.0923498347401619,
|
|
"num_tokens": 577084.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.309446048736572,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.972,
|
|
"mean_token_accuracy": 0.09742946326732635,
|
|
"num_tokens": 587980.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.299356460571289,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8505,
|
|
"mean_token_accuracy": 0.10278590098023414,
|
|
"num_tokens": 598989.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.4622321128845215,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0462,
|
|
"mean_token_accuracy": 0.09367327913641929,
|
|
"num_tokens": 608599.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.311740112304688,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.09894169121980667,
|
|
"num_tokens": 619267.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.329383373260498,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0237,
|
|
"mean_token_accuracy": 0.0995009183883667,
|
|
"num_tokens": 629991.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.334739446640015,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9751,
|
|
"mean_token_accuracy": 0.09126680493354797,
|
|
"num_tokens": 641311.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.340094423294067,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9349,
|
|
"mean_token_accuracy": 0.0958991676568985,
|
|
"num_tokens": 651812.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.323620367050171,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.8849,
|
|
"mean_token_accuracy": 0.10023088753223419,
|
|
"num_tokens": 662387.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.2562236309051515,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8697,
|
|
"mean_token_accuracy": 0.09553462415933608,
|
|
"num_tokens": 673018.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.216851282119751,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8094,
|
|
"mean_token_accuracy": 0.09729016497731209,
|
|
"num_tokens": 683192.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.253824520111084,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9097,
|
|
"mean_token_accuracy": 0.10488839745521546,
|
|
"num_tokens": 692747.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.1223766803741455,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.09245335534214974,
|
|
"num_tokens": 705109.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.2870441436767575,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8193,
|
|
"mean_token_accuracy": 0.09904048666357994,
|
|
"num_tokens": 715658.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.1810273170471195,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.8301,
|
|
"mean_token_accuracy": 0.10451033040881157,
|
|
"num_tokens": 724961.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.068193769454956,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7801,
|
|
"mean_token_accuracy": 0.10440796837210656,
|
|
"num_tokens": 735416.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.2093346118927,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8719,
|
|
"mean_token_accuracy": 0.10421840399503708,
|
|
"num_tokens": 745005.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.2864179611206055,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9046,
|
|
"mean_token_accuracy": 0.10184969678521157,
|
|
"num_tokens": 754743.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.0180971145629885,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7603,
|
|
"mean_token_accuracy": 0.10306568518280983,
|
|
"num_tokens": 764897.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.152372789382935,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8267,
|
|
"mean_token_accuracy": 0.09857897013425827,
|
|
"num_tokens": 775302.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.124710941314698,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7588,
|
|
"mean_token_accuracy": 0.10548576042056083,
|
|
"num_tokens": 785694.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.054194736480713,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.72,
|
|
"mean_token_accuracy": 0.11008026450872421,
|
|
"num_tokens": 795045.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.09068751335144,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.781,
|
|
"mean_token_accuracy": 0.10049664527177811,
|
|
"num_tokens": 805579.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.095667552947998,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.6838,
|
|
"mean_token_accuracy": 0.10740630999207497,
|
|
"num_tokens": 816743.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.093503808975219,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8234,
|
|
"mean_token_accuracy": 0.09930850043892861,
|
|
"num_tokens": 827217.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.012804555892944,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6731,
|
|
"mean_token_accuracy": 0.10424429401755334,
|
|
"num_tokens": 837555.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.039196920394898,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7459,
|
|
"mean_token_accuracy": 0.10508461520075799,
|
|
"num_tokens": 846726.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.120304107666016,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7475,
|
|
"mean_token_accuracy": 0.10223370566964149,
|
|
"num_tokens": 856535.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.005828332901001,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.8069,
|
|
"mean_token_accuracy": 0.09859322756528854,
|
|
"num_tokens": 866973.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.085656690597534,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7415,
|
|
"mean_token_accuracy": 0.1009980633854866,
|
|
"num_tokens": 877873.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.018812036514282,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7335,
|
|
"mean_token_accuracy": 0.10494705140590668,
|
|
"num_tokens": 887135.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.096279811859131,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7514,
|
|
"mean_token_accuracy": 0.10019353330135346,
|
|
"num_tokens": 897206.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.024485397338867,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7353,
|
|
"mean_token_accuracy": 0.10563584491610527,
|
|
"num_tokens": 906808.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.898047828674317,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.6422,
|
|
"mean_token_accuracy": 0.11458711996674538,
|
|
"num_tokens": 917177.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.959504508972168,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6907,
|
|
"mean_token_accuracy": 0.10576008260250092,
|
|
"num_tokens": 927229.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.006360816955566,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6812,
|
|
"mean_token_accuracy": 0.11021844074130058,
|
|
"num_tokens": 937832.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.870974111557007,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6064,
|
|
"mean_token_accuracy": 0.10718697234988213,
|
|
"num_tokens": 948038.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.002683067321778,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7181,
|
|
"mean_token_accuracy": 0.10097691044211388,
|
|
"num_tokens": 957924.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.933505868911743,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6792,
|
|
"mean_token_accuracy": 0.10600680485367775,
|
|
"num_tokens": 968497.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.915420055389404,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6043,
|
|
"mean_token_accuracy": 0.11050590574741363,
|
|
"num_tokens": 978878.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.828244590759278,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7024,
|
|
"mean_token_accuracy": 0.10729977861046791,
|
|
"num_tokens": 989984.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.974292421340943,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.6266,
|
|
"mean_token_accuracy": 0.1099898062646389,
|
|
"num_tokens": 1000254.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.815394496917724,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.687,
|
|
"mean_token_accuracy": 0.10523343607783317,
|
|
"num_tokens": 1010697.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.067700672149658,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7775,
|
|
"mean_token_accuracy": 0.10482538044452668,
|
|
"num_tokens": 1020892.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.969514179229736,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6552,
|
|
"mean_token_accuracy": 0.11519927084445954,
|
|
"num_tokens": 1030380.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 6.938153394432955,
|
|
"eval_loss": 6.763866901397705,
|
|
"eval_mean_token_accuracy": 0.10726368882974913,
|
|
"eval_num_tokens": 1030380.0,
|
|
"eval_runtime": 2.6364,
|
|
"eval_samples_per_second": 1302.937,
|
|
"eval_steps_per_second": 163.104,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.912630319595337,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7031,
|
|
"mean_token_accuracy": 0.10474735051393509,
|
|
"num_tokens": 1040301.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.821575069427491,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.5677,
|
|
"mean_token_accuracy": 0.11079607084393502,
|
|
"num_tokens": 1051999.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.036813116073608,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7879,
|
|
"mean_token_accuracy": 0.10318484753370286,
|
|
"num_tokens": 1063133.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.810295963287354,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6059,
|
|
"mean_token_accuracy": 0.10231919735670089,
|
|
"num_tokens": 1073972.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.8240800380706785,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5935,
|
|
"mean_token_accuracy": 0.10545961037278176,
|
|
"num_tokens": 1084050.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.935742712020874,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7033,
|
|
"mean_token_accuracy": 0.1093459077179432,
|
|
"num_tokens": 1094415.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.71245813369751,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6194,
|
|
"mean_token_accuracy": 0.10769399031996726,
|
|
"num_tokens": 1105133.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.876086091995239,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.5726,
|
|
"mean_token_accuracy": 0.11468547657132148,
|
|
"num_tokens": 1115266.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.729256963729858,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.5545,
|
|
"mean_token_accuracy": 0.1140455037355423,
|
|
"num_tokens": 1124775.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.831462669372558,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6004,
|
|
"mean_token_accuracy": 0.11026408672332763,
|
|
"num_tokens": 1134498.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.807862949371338,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6423,
|
|
"mean_token_accuracy": 0.10805548503994941,
|
|
"num_tokens": 1146067.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.918079137802124,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6528,
|
|
"mean_token_accuracy": 0.10948517173528671,
|
|
"num_tokens": 1156349.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.835301733016967,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.5719,
|
|
"mean_token_accuracy": 0.10832828059792518,
|
|
"num_tokens": 1166851.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.76150369644165,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.527,
|
|
"mean_token_accuracy": 0.11024561673402786,
|
|
"num_tokens": 1176899.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.7258953094482425,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.6245,
|
|
"mean_token_accuracy": 0.10835396870970726,
|
|
"num_tokens": 1187383.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.841786193847656,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.601,
|
|
"mean_token_accuracy": 0.11034611463546753,
|
|
"num_tokens": 1197836.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.7962751388549805,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.5547,
|
|
"mean_token_accuracy": 0.1168977290391922,
|
|
"num_tokens": 1208533.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.752965354919434,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.5959,
|
|
"mean_token_accuracy": 0.10688599348068237,
|
|
"num_tokens": 1219397.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.719386243820191,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.5847,
|
|
"mean_token_accuracy": 0.10793909430503845,
|
|
"num_tokens": 1229991.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.685470914840698,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.487,
|
|
"mean_token_accuracy": 0.10784793198108673,
|
|
"num_tokens": 1240343.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.747214031219483,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.5528,
|
|
"mean_token_accuracy": 0.10783586576581002,
|
|
"num_tokens": 1250432.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.683514261245728,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4859,
|
|
"mean_token_accuracy": 0.11751269921660423,
|
|
"num_tokens": 1260744.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.735761070251465,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.516,
|
|
"mean_token_accuracy": 0.11311889961361885,
|
|
"num_tokens": 1272155.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.594959259033203,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3957,
|
|
"mean_token_accuracy": 0.11406432464718819,
|
|
"num_tokens": 1282240.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.700835132598877,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5768,
|
|
"mean_token_accuracy": 0.11421498507261277,
|
|
"num_tokens": 1292169.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.643769598007202,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.474,
|
|
"mean_token_accuracy": 0.11547293737530709,
|
|
"num_tokens": 1302010.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.698951387405396,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.5299,
|
|
"mean_token_accuracy": 0.11720511168241501,
|
|
"num_tokens": 1312596.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.670014810562134,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4806,
|
|
"mean_token_accuracy": 0.10972161293029785,
|
|
"num_tokens": 1321753.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.6802314758300785,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.5269,
|
|
"mean_token_accuracy": 0.11095310598611832,
|
|
"num_tokens": 1332059.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.659747362136841,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.5006,
|
|
"mean_token_accuracy": 0.11720726266503334,
|
|
"num_tokens": 1341517.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.768359041213989,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5744,
|
|
"mean_token_accuracy": 0.11781106889247894,
|
|
"num_tokens": 1351439.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.605283737182617,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4471,
|
|
"mean_token_accuracy": 0.11385733485221863,
|
|
"num_tokens": 1360745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.614739894866943,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.4547,
|
|
"mean_token_accuracy": 0.11523936912417412,
|
|
"num_tokens": 1371931.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.524019193649292,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4051,
|
|
"mean_token_accuracy": 0.11907100528478623,
|
|
"num_tokens": 1382040.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.583529663085938,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4651,
|
|
"mean_token_accuracy": 0.10969016775488853,
|
|
"num_tokens": 1392690.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.671084403991699,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.4711,
|
|
"mean_token_accuracy": 0.11799429208040238,
|
|
"num_tokens": 1403060.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.62229208946228,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.4732,
|
|
"mean_token_accuracy": 0.11298275738954544,
|
|
"num_tokens": 1412895.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.59600863456726,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.425,
|
|
"mean_token_accuracy": 0.12277772873640061,
|
|
"num_tokens": 1423024.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.580408048629761,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.4492,
|
|
"mean_token_accuracy": 0.11523076891899109,
|
|
"num_tokens": 1433578.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.551895475387573,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4945,
|
|
"mean_token_accuracy": 0.11739786639809609,
|
|
"num_tokens": 1444473.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.677528762817383,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.513,
|
|
"mean_token_accuracy": 0.11457236558198929,
|
|
"num_tokens": 1454553.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.594537878036499,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.368,
|
|
"mean_token_accuracy": 0.11736505180597305,
|
|
"num_tokens": 1464162.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.536549663543701,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.11977096050977706,
|
|
"num_tokens": 1474691.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.626703071594238,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5126,
|
|
"mean_token_accuracy": 0.11911239996552467,
|
|
"num_tokens": 1484650.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.5932032585144045,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4708,
|
|
"mean_token_accuracy": 0.11753673776984215,
|
|
"num_tokens": 1494788.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.577540493011474,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.4192,
|
|
"mean_token_accuracy": 0.11684553176164628,
|
|
"num_tokens": 1506574.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.492519283294678,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.44,
|
|
"mean_token_accuracy": 0.11481376960873604,
|
|
"num_tokens": 1517505.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.5567238330841064,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3389,
|
|
"mean_token_accuracy": 0.12622489854693414,
|
|
"num_tokens": 1526662.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.474819231033325,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4503,
|
|
"mean_token_accuracy": 0.117062958329916,
|
|
"num_tokens": 1536798.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.589357805252075,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.3611,
|
|
"mean_token_accuracy": 0.12048957645893096,
|
|
"num_tokens": 1546673.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.47542667388916,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.3815,
|
|
"mean_token_accuracy": 0.12386295571923256,
|
|
"num_tokens": 1556793.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.547279596328735,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4267,
|
|
"mean_token_accuracy": 0.12142022401094436,
|
|
"num_tokens": 1566450.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.490754413604736,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4515,
|
|
"mean_token_accuracy": 0.12024893313646316,
|
|
"num_tokens": 1576353.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.525695848464966,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3332,
|
|
"mean_token_accuracy": 0.12352794036269188,
|
|
"num_tokens": 1586318.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.536564779281616,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4345,
|
|
"mean_token_accuracy": 0.11370632126927376,
|
|
"num_tokens": 1596839.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.4864342212677,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4529,
|
|
"mean_token_accuracy": 0.12249588519334793,
|
|
"num_tokens": 1606843.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.363466262817383,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4148,
|
|
"mean_token_accuracy": 0.11689037606120109,
|
|
"num_tokens": 1617048.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.529700660705567,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.3213,
|
|
"mean_token_accuracy": 0.12030550986528396,
|
|
"num_tokens": 1627472.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.547754430770874,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.429,
|
|
"mean_token_accuracy": 0.11561316847801209,
|
|
"num_tokens": 1637716.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.427017641067505,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4385,
|
|
"mean_token_accuracy": 0.11526587828993798,
|
|
"num_tokens": 1648119.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.5540087699890135,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3766,
|
|
"mean_token_accuracy": 0.1245071604847908,
|
|
"num_tokens": 1658214.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.3835773944854735,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3437,
|
|
"mean_token_accuracy": 0.12322835624217987,
|
|
"num_tokens": 1668725.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.435897397994995,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4166,
|
|
"mean_token_accuracy": 0.12298030406236649,
|
|
"num_tokens": 1678891.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.519479846954345,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.3176,
|
|
"mean_token_accuracy": 0.12078626230359077,
|
|
"num_tokens": 1688749.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.44858684539795,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4246,
|
|
"mean_token_accuracy": 0.11904875412583352,
|
|
"num_tokens": 1699166.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.509427452087403,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.348,
|
|
"mean_token_accuracy": 0.12165467068552971,
|
|
"num_tokens": 1710690.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.475676584243774,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4505,
|
|
"mean_token_accuracy": 0.11575704216957092,
|
|
"num_tokens": 1721169.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.419601774215698,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3168,
|
|
"mean_token_accuracy": 0.11946589276194572,
|
|
"num_tokens": 1731523.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.477061462402344,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.3088,
|
|
"mean_token_accuracy": 0.11968929320573807,
|
|
"num_tokens": 1742316.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.44489312171936,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4524,
|
|
"mean_token_accuracy": 0.12095453590154648,
|
|
"num_tokens": 1753424.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.338519668579101,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2815,
|
|
"mean_token_accuracy": 0.12474420070648193,
|
|
"num_tokens": 1762814.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.428976821899414,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4394,
|
|
"mean_token_accuracy": 0.11995334699749946,
|
|
"num_tokens": 1773096.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.479230260848999,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.12182644307613373,
|
|
"num_tokens": 1783499.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.607985353469848,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.5456,
|
|
"mean_token_accuracy": 0.1136684074997902,
|
|
"num_tokens": 1794107.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.4030835151672365,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4216,
|
|
"mean_token_accuracy": 0.1149469256401062,
|
|
"num_tokens": 1805151.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.488826656341553,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.391,
|
|
"mean_token_accuracy": 0.11763054057955742,
|
|
"num_tokens": 1815205.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.399917316436768,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.3587,
|
|
"mean_token_accuracy": 0.1239665150642395,
|
|
"num_tokens": 1825301.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.424862003326416,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.379,
|
|
"mean_token_accuracy": 0.12152353450655937,
|
|
"num_tokens": 1835634.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.472187471389771,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3892,
|
|
"mean_token_accuracy": 0.11498059630393982,
|
|
"num_tokens": 1846855.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.382741403579712,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3587,
|
|
"mean_token_accuracy": 0.12044938951730728,
|
|
"num_tokens": 1857747.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.446867799758911,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.4105,
|
|
"mean_token_accuracy": 0.12059052139520646,
|
|
"num_tokens": 1868098.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.350548887252808,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3093,
|
|
"mean_token_accuracy": 0.12493439465761184,
|
|
"num_tokens": 1879682.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.432272911071777,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.3316,
|
|
"mean_token_accuracy": 0.12454040497541427,
|
|
"num_tokens": 1890605.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.39534969329834,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.3699,
|
|
"mean_token_accuracy": 0.12763101756572723,
|
|
"num_tokens": 1900861.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.398876667022705,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3464,
|
|
"mean_token_accuracy": 0.12193151786923409,
|
|
"num_tokens": 1910867.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.434603929519653,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3612,
|
|
"mean_token_accuracy": 0.11605243608355523,
|
|
"num_tokens": 1921424.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.368513393402099,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.3433,
|
|
"mean_token_accuracy": 0.1275160454213619,
|
|
"num_tokens": 1930784.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.395671701431274,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3324,
|
|
"mean_token_accuracy": 0.12198550030589103,
|
|
"num_tokens": 1940325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.335153245925904,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3475,
|
|
"mean_token_accuracy": 0.12423227652907372,
|
|
"num_tokens": 1951799.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.4373150825500485,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.11780767813324929,
|
|
"num_tokens": 1962311.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.351272106170654,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.338,
|
|
"mean_token_accuracy": 0.12228847667574883,
|
|
"num_tokens": 1974128.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.251103639602661,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.12811544090509414,
|
|
"num_tokens": 1983765.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.386243391036987,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3339,
|
|
"mean_token_accuracy": 0.1216882549226284,
|
|
"num_tokens": 1994045.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.403064489364624,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4084,
|
|
"mean_token_accuracy": 0.11930646523833274,
|
|
"num_tokens": 2005353.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.219151544570923,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.2239,
|
|
"mean_token_accuracy": 0.12836714312434197,
|
|
"num_tokens": 2015808.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.362400388717651,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2823,
|
|
"mean_token_accuracy": 0.1286115452647209,
|
|
"num_tokens": 2025824.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.220054244995117,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.2977,
|
|
"mean_token_accuracy": 0.1320819690823555,
|
|
"num_tokens": 2036924.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.383143568038941,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3253,
|
|
"mean_token_accuracy": 0.12373943105340005,
|
|
"num_tokens": 2047302.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.269778537750244,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3067,
|
|
"mean_token_accuracy": 0.12240820974111558,
|
|
"num_tokens": 2057972.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.2414275169372555,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.154,
|
|
"mean_token_accuracy": 0.12995185926556588,
|
|
"num_tokens": 2067916.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.18265976351361,
|
|
"eval_loss": 6.422311782836914,
|
|
"eval_mean_token_accuracy": 0.12340358548559421,
|
|
"eval_num_tokens": 2067916.0,
|
|
"eval_runtime": 2.5935,
|
|
"eval_samples_per_second": 1324.482,
|
|
"eval_steps_per_second": 165.801,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.2358527183532715,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.3354,
|
|
"mean_token_accuracy": 0.1269114390015602,
|
|
"num_tokens": 2079119.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.302850770950317,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.224,
|
|
"mean_token_accuracy": 0.12514718621969223,
|
|
"num_tokens": 2090080.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.266329765319824,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.2681,
|
|
"mean_token_accuracy": 0.12983603328466414,
|
|
"num_tokens": 2100541.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.328331470489502,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.2368,
|
|
"mean_token_accuracy": 0.12401319071650505,
|
|
"num_tokens": 2111213.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.298262405395508,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.3306,
|
|
"mean_token_accuracy": 0.12413369193673134,
|
|
"num_tokens": 2121714.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.274141407012939,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.2165,
|
|
"mean_token_accuracy": 0.12441339045763015,
|
|
"num_tokens": 2131226.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.270669841766358,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.2877,
|
|
"mean_token_accuracy": 0.11828686073422431,
|
|
"num_tokens": 2143179.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.220635652542114,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.12587330713868142,
|
|
"num_tokens": 2153646.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.198286867141723,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.2264,
|
|
"mean_token_accuracy": 0.12276902943849563,
|
|
"num_tokens": 2163267.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.2652397632598875,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.1411,
|
|
"mean_token_accuracy": 0.1308826483786106,
|
|
"num_tokens": 2173282.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.211602592468262,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.2276,
|
|
"mean_token_accuracy": 0.12286282628774643,
|
|
"num_tokens": 2184202.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.3284876346588135,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.2028,
|
|
"mean_token_accuracy": 0.12903301566839218,
|
|
"num_tokens": 2194342.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.160139703750611,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.198,
|
|
"mean_token_accuracy": 0.1318102538585663,
|
|
"num_tokens": 2204817.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.2580382347106935,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.2727,
|
|
"mean_token_accuracy": 0.13099218532443047,
|
|
"num_tokens": 2215281.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.3193916320800785,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.2859,
|
|
"mean_token_accuracy": 0.1294151432812214,
|
|
"num_tokens": 2225784.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.138168430328369,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.2205,
|
|
"mean_token_accuracy": 0.1334165170788765,
|
|
"num_tokens": 2235357.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.378276205062866,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.2359,
|
|
"mean_token_accuracy": 0.13057358041405678,
|
|
"num_tokens": 2246438.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.202377271652222,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.2261,
|
|
"mean_token_accuracy": 0.12415477558970452,
|
|
"num_tokens": 2257528.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.155335521697998,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.2498,
|
|
"mean_token_accuracy": 0.13259016051888467,
|
|
"num_tokens": 2267572.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.226053190231323,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.1789,
|
|
"mean_token_accuracy": 0.13369767963886262,
|
|
"num_tokens": 2278520.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.2539831267462835,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 6.0826,
|
|
"mean_token_accuracy": 0.1368540914522277,
|
|
"num_tokens": 2286991.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.170029067993164,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 6.1303,
|
|
"mean_token_accuracy": 0.12599217817187308,
|
|
"num_tokens": 2297857.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.261499547958374,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 6.1702,
|
|
"mean_token_accuracy": 0.13075068444013596,
|
|
"num_tokens": 2308621.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.14419322013855,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 5.9933,
|
|
"mean_token_accuracy": 0.14160198718309402,
|
|
"num_tokens": 2318654.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.1928181648254395,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 6.045,
|
|
"mean_token_accuracy": 0.1332675762474537,
|
|
"num_tokens": 2329064.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.150006675720215,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 6.1098,
|
|
"mean_token_accuracy": 0.12820298597216606,
|
|
"num_tokens": 2339843.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.200477075576782,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 6.0436,
|
|
"mean_token_accuracy": 0.13033792674541472,
|
|
"num_tokens": 2350715.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.130117893218994,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 6.0068,
|
|
"mean_token_accuracy": 0.13114059641957282,
|
|
"num_tokens": 2360635.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.180344820022583,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 6.0305,
|
|
"mean_token_accuracy": 0.13138258829712868,
|
|
"num_tokens": 2370627.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.147628831863403,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 6.0424,
|
|
"mean_token_accuracy": 0.13276968747377396,
|
|
"num_tokens": 2381790.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.058908224105835,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.13258678764104842,
|
|
"num_tokens": 2392685.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.140621614456177,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 6.0486,
|
|
"mean_token_accuracy": 0.13231537640094757,
|
|
"num_tokens": 2403722.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.107622337341309,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.0933,
|
|
"mean_token_accuracy": 0.12645822986960412,
|
|
"num_tokens": 2415311.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.148067760467529,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 5.9773,
|
|
"mean_token_accuracy": 0.13930070772767067,
|
|
"num_tokens": 2425116.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.2148473262786865,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 6.1548,
|
|
"mean_token_accuracy": 0.1310862734913826,
|
|
"num_tokens": 2435916.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.043871068954468,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 6.0384,
|
|
"mean_token_accuracy": 0.12819197103381158,
|
|
"num_tokens": 2446920.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.161314392089844,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 5.9899,
|
|
"mean_token_accuracy": 0.1420701712369919,
|
|
"num_tokens": 2456367.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.064000463485717,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.13507229685783387,
|
|
"num_tokens": 2466790.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.089735412597657,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 5.968,
|
|
"mean_token_accuracy": 0.1441962629556656,
|
|
"num_tokens": 2476419.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.057880926132202,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 6.1084,
|
|
"mean_token_accuracy": 0.13047450110316278,
|
|
"num_tokens": 2487295.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.063084745407105,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 5.9996,
|
|
"mean_token_accuracy": 0.12964291498064995,
|
|
"num_tokens": 2497545.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.217773675918579,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 6.1494,
|
|
"mean_token_accuracy": 0.12842334657907487,
|
|
"num_tokens": 2507805.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.087493848800659,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 6.0238,
|
|
"mean_token_accuracy": 0.13358174711465837,
|
|
"num_tokens": 2518474.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.1921933650970455,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 6.047,
|
|
"mean_token_accuracy": 0.14280006289482117,
|
|
"num_tokens": 2528974.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.036006498336792,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 6.0951,
|
|
"mean_token_accuracy": 0.12937285155057907,
|
|
"num_tokens": 2540675.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.117119979858399,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 5.9689,
|
|
"mean_token_accuracy": 0.1407436214387417,
|
|
"num_tokens": 2551367.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.063678407669068,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 6.062,
|
|
"mean_token_accuracy": 0.13190347701311111,
|
|
"num_tokens": 2562096.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.158404207229614,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 6.1277,
|
|
"mean_token_accuracy": 0.13364789113402367,
|
|
"num_tokens": 2572917.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.0960992813110355,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 5.9614,
|
|
"mean_token_accuracy": 0.1394283317029476,
|
|
"num_tokens": 2582971.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.16928243637085,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 6.1091,
|
|
"mean_token_accuracy": 0.13041222244501113,
|
|
"num_tokens": 2593186.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.133707523345947,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 6.0997,
|
|
"mean_token_accuracy": 0.13235297948122024,
|
|
"num_tokens": 2604359.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.136443281173706,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.13984801024198532,
|
|
"num_tokens": 2614436.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.983046245574951,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 5.9837,
|
|
"mean_token_accuracy": 0.13748598098754883,
|
|
"num_tokens": 2624175.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.080909395217896,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 6.0246,
|
|
"mean_token_accuracy": 0.13808856904506683,
|
|
"num_tokens": 2634898.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.203506231307983,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 6.0522,
|
|
"mean_token_accuracy": 0.13835761845111846,
|
|
"num_tokens": 2645590.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.1103081703186035,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 6.039,
|
|
"mean_token_accuracy": 0.13795397132635118,
|
|
"num_tokens": 2656115.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.081531190872193,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 6.0248,
|
|
"mean_token_accuracy": 0.1361389346420765,
|
|
"num_tokens": 2666760.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.077518939971924,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.13790025189518929,
|
|
"num_tokens": 2676763.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.0636731624603275,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 6.0108,
|
|
"mean_token_accuracy": 0.13075025081634523,
|
|
"num_tokens": 2686598.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.058450698852539,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 5.9813,
|
|
"mean_token_accuracy": 0.13843700140714646,
|
|
"num_tokens": 2697767.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.048785066604614,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 5.9743,
|
|
"mean_token_accuracy": 0.1362238995730877,
|
|
"num_tokens": 2707081.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.002673721313476,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 6.0005,
|
|
"mean_token_accuracy": 0.1412185825407505,
|
|
"num_tokens": 2717312.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.1222405433654785,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 6.0165,
|
|
"mean_token_accuracy": 0.1373762771487236,
|
|
"num_tokens": 2728108.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.153705739974976,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 6.2116,
|
|
"mean_token_accuracy": 0.1348584771156311,
|
|
"num_tokens": 2738860.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.08194637298584,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.14144043251872063,
|
|
"num_tokens": 2749786.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.067605495452881,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.9576,
|
|
"mean_token_accuracy": 0.1407739117741585,
|
|
"num_tokens": 2760018.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.014369344711303,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 5.9924,
|
|
"mean_token_accuracy": 0.13832346946001053,
|
|
"num_tokens": 2770572.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.000287675857544,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.9557,
|
|
"mean_token_accuracy": 0.14493367075920105,
|
|
"num_tokens": 2780297.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.126511573791504,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 6.0754,
|
|
"mean_token_accuracy": 0.1383805565536022,
|
|
"num_tokens": 2790454.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.051705694198608,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 5.9269,
|
|
"mean_token_accuracy": 0.14142982065677642,
|
|
"num_tokens": 2800034.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.937855005264282,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 5.9515,
|
|
"mean_token_accuracy": 0.14328320026397706,
|
|
"num_tokens": 2810157.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.0591309547424315,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.9721,
|
|
"mean_token_accuracy": 0.1326573297381401,
|
|
"num_tokens": 2819404.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.9855554580688475,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.912,
|
|
"mean_token_accuracy": 0.13907658755779267,
|
|
"num_tokens": 2829551.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.990193843841553,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 5.9849,
|
|
"mean_token_accuracy": 0.1325812466442585,
|
|
"num_tokens": 2840381.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.0108177185058596,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 5.9657,
|
|
"mean_token_accuracy": 0.14018113762140275,
|
|
"num_tokens": 2850279.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.081850671768189,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 5.999,
|
|
"mean_token_accuracy": 0.1427960455417633,
|
|
"num_tokens": 2860209.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.986414766311645,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 6.0065,
|
|
"mean_token_accuracy": 0.1422063134610653,
|
|
"num_tokens": 2870932.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.154019975662232,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.13413980305194856,
|
|
"num_tokens": 2881139.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.022779893875122,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.9898,
|
|
"mean_token_accuracy": 0.13735412806272507,
|
|
"num_tokens": 2890774.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.014951515197754,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 5.9483,
|
|
"mean_token_accuracy": 0.14736960977315902,
|
|
"num_tokens": 2900259.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.024165916442871,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.14006007090210915,
|
|
"num_tokens": 2911365.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.027864694595337,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.9561,
|
|
"mean_token_accuracy": 0.13755545094609262,
|
|
"num_tokens": 2921189.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.897766399383545,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.9868,
|
|
"mean_token_accuracy": 0.1411122649908066,
|
|
"num_tokens": 2931271.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.041615962982178,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.9812,
|
|
"mean_token_accuracy": 0.13669976145029067,
|
|
"num_tokens": 2941557.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.959131717681885,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 5.8968,
|
|
"mean_token_accuracy": 0.14948886036872863,
|
|
"num_tokens": 2951305.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.962688350677491,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 5.9791,
|
|
"mean_token_accuracy": 0.14904043525457383,
|
|
"num_tokens": 2961948.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.03700361251831,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 6.0435,
|
|
"mean_token_accuracy": 0.1346405990421772,
|
|
"num_tokens": 2973300.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.081942367553711,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 5.9856,
|
|
"mean_token_accuracy": 0.13874733671545983,
|
|
"num_tokens": 2984114.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.880176210403443,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.8527,
|
|
"mean_token_accuracy": 0.1486996442079544,
|
|
"num_tokens": 2993940.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.036293983459473,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.1445149928331375,
|
|
"num_tokens": 3004093.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.859427356719971,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.894,
|
|
"mean_token_accuracy": 0.14320220649242402,
|
|
"num_tokens": 3013537.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.090070915222168,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.8986,
|
|
"mean_token_accuracy": 0.14571946561336518,
|
|
"num_tokens": 3023502.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.919493579864502,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.13945504799485206,
|
|
"num_tokens": 3033910.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.9833661079406735,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 5.9951,
|
|
"mean_token_accuracy": 0.14348382130265236,
|
|
"num_tokens": 3043500.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.971828985214233,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.9641,
|
|
"mean_token_accuracy": 0.14203015938401223,
|
|
"num_tokens": 3054776.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.034963607788086,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.9249,
|
|
"mean_token_accuracy": 0.14127751886844636,
|
|
"num_tokens": 3065403.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.924037456512451,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 5.9545,
|
|
"mean_token_accuracy": 0.14617739617824554,
|
|
"num_tokens": 3074661.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.930827283859253,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.9193,
|
|
"mean_token_accuracy": 0.13931230902671815,
|
|
"num_tokens": 3085319.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.930323791503906,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.9068,
|
|
"mean_token_accuracy": 0.1437447987496853,
|
|
"num_tokens": 3095115.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.019172143936157,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 6.0843,
|
|
"mean_token_accuracy": 0.13452800139784812,
|
|
"num_tokens": 3104841.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 5.952520974846774,
|
|
"eval_loss": 6.207791805267334,
|
|
"eval_mean_token_accuracy": 0.13844329656729865,
|
|
"eval_num_tokens": 3104841.0,
|
|
"eval_runtime": 2.7946,
|
|
"eval_samples_per_second": 1229.15,
|
|
"eval_steps_per_second": 153.867,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.99462194442749,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 5.9409,
|
|
"mean_token_accuracy": 0.14266305193305015,
|
|
"num_tokens": 3115098.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.979383325576782,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 5.8424,
|
|
"mean_token_accuracy": 0.15014744699001312,
|
|
"num_tokens": 3124831.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.916868686676025,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.13746115788817406,
|
|
"num_tokens": 3136113.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.122059392929077,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.9268,
|
|
"mean_token_accuracy": 0.15368309766054153,
|
|
"num_tokens": 3145617.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.852825069427491,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 6.0147,
|
|
"mean_token_accuracy": 0.14040548503398895,
|
|
"num_tokens": 3156913.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.042289066314697,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.9586,
|
|
"mean_token_accuracy": 0.13911829963326455,
|
|
"num_tokens": 3167506.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.871926736831665,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.8468,
|
|
"mean_token_accuracy": 0.14407547563314438,
|
|
"num_tokens": 3177956.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.061702823638916,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 6.0998,
|
|
"mean_token_accuracy": 0.14125556200742723,
|
|
"num_tokens": 3189172.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.007503175735474,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 6.0076,
|
|
"mean_token_accuracy": 0.1400210067629814,
|
|
"num_tokens": 3200188.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.946048545837402,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.8475,
|
|
"mean_token_accuracy": 0.14049874618649483,
|
|
"num_tokens": 3210171.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.00736141204834,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 6.0701,
|
|
"mean_token_accuracy": 0.13583848625421524,
|
|
"num_tokens": 3221201.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.953229236602783,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 5.921,
|
|
"mean_token_accuracy": 0.1424741789698601,
|
|
"num_tokens": 3231825.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.091651391983032,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 6.114,
|
|
"mean_token_accuracy": 0.13234196305274964,
|
|
"num_tokens": 3241860.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.877058601379394,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.7883,
|
|
"mean_token_accuracy": 0.15056520849466323,
|
|
"num_tokens": 3251893.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.967958641052246,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.9205,
|
|
"mean_token_accuracy": 0.13848666921257974,
|
|
"num_tokens": 3262748.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.038754320144653,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 6.0288,
|
|
"mean_token_accuracy": 0.13997522443532945,
|
|
"num_tokens": 3273035.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.02143964767456,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.9559,
|
|
"mean_token_accuracy": 0.14435344338417053,
|
|
"num_tokens": 3283413.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.952090311050415,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.1430325210094452,
|
|
"num_tokens": 3294628.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.941321563720703,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.15504622757434844,
|
|
"num_tokens": 3304631.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.942373561859131,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 5.9472,
|
|
"mean_token_accuracy": 0.14010237380862237,
|
|
"num_tokens": 3315481.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.963896894454956,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.8896,
|
|
"mean_token_accuracy": 0.14427914768457412,
|
|
"num_tokens": 3326493.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.1022755146026615,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 6.0348,
|
|
"mean_token_accuracy": 0.1388540171086788,
|
|
"num_tokens": 3338358.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.969568681716919,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.9892,
|
|
"mean_token_accuracy": 0.1384444311261177,
|
|
"num_tokens": 3347961.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.955298995971679,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.918,
|
|
"mean_token_accuracy": 0.14495049715042113,
|
|
"num_tokens": 3358544.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.940098476409912,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.14714469611644745,
|
|
"num_tokens": 3368233.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.891028165817261,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.8798,
|
|
"mean_token_accuracy": 0.14687413945794106,
|
|
"num_tokens": 3379065.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.050700235366821,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.9705,
|
|
"mean_token_accuracy": 0.1408476561307907,
|
|
"num_tokens": 3389251.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.947952032089233,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.904,
|
|
"mean_token_accuracy": 0.14892102256417275,
|
|
"num_tokens": 3399660.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.9753376007080075,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.952,
|
|
"mean_token_accuracy": 0.14645659551024437,
|
|
"num_tokens": 3409395.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.021202087402344,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.9398,
|
|
"mean_token_accuracy": 0.136006910353899,
|
|
"num_tokens": 3419881.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.926182079315185,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.9319,
|
|
"mean_token_accuracy": 0.14269742593169213,
|
|
"num_tokens": 3430316.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.884389495849609,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.8264,
|
|
"mean_token_accuracy": 0.15141425877809525,
|
|
"num_tokens": 3440174.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.954122495651245,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.9859,
|
|
"mean_token_accuracy": 0.13919273614883423,
|
|
"num_tokens": 3451141.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.008854818344116,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.8835,
|
|
"mean_token_accuracy": 0.14436262398958205,
|
|
"num_tokens": 3461767.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.854226875305176,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.15278302133083344,
|
|
"num_tokens": 3470937.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.9624449729919435,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.9211,
|
|
"mean_token_accuracy": 0.13923487737774848,
|
|
"num_tokens": 3482130.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.961441612243652,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.9873,
|
|
"mean_token_accuracy": 0.14150320515036582,
|
|
"num_tokens": 3491992.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.8464337348937985,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.844,
|
|
"mean_token_accuracy": 0.15078289955854415,
|
|
"num_tokens": 3501437.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.876981639862061,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.9118,
|
|
"mean_token_accuracy": 0.14388926029205323,
|
|
"num_tokens": 3511625.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.918456602096557,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 5.7004,
|
|
"mean_token_accuracy": 0.15261678099632264,
|
|
"num_tokens": 3520428.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.920832633972168,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.9139,
|
|
"mean_token_accuracy": 0.14273062944412232,
|
|
"num_tokens": 3532218.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.941348266601563,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.8974,
|
|
"mean_token_accuracy": 0.14131915792822838,
|
|
"num_tokens": 3543177.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.9505699634552,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.8494,
|
|
"mean_token_accuracy": 0.1531837522983551,
|
|
"num_tokens": 3552833.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.805274772644043,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.9233,
|
|
"mean_token_accuracy": 0.1416128844022751,
|
|
"num_tokens": 3563558.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.891717195510864,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.8716,
|
|
"mean_token_accuracy": 0.15252673625946045,
|
|
"num_tokens": 3574993.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.977005434036255,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.89,
|
|
"mean_token_accuracy": 0.14125805348157883,
|
|
"num_tokens": 3584586.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.831153774261475,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 5.8367,
|
|
"mean_token_accuracy": 0.15189587622880935,
|
|
"num_tokens": 3595029.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.910392475128174,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.8994,
|
|
"mean_token_accuracy": 0.145536919683218,
|
|
"num_tokens": 3606034.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.950764417648315,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.8402,
|
|
"mean_token_accuracy": 0.14515646025538445,
|
|
"num_tokens": 3615551.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.84814076423645,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.9831,
|
|
"mean_token_accuracy": 0.1476181536912918,
|
|
"num_tokens": 3626811.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.922665405273437,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.8358,
|
|
"mean_token_accuracy": 0.14447919875383378,
|
|
"num_tokens": 3637338.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.839316034317017,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.7726,
|
|
"mean_token_accuracy": 0.14876675009727477,
|
|
"num_tokens": 3648650.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.943630695343018,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.9341,
|
|
"mean_token_accuracy": 0.14553604349493982,
|
|
"num_tokens": 3659331.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.960559415817261,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 6.0046,
|
|
"mean_token_accuracy": 0.13882950097322463,
|
|
"num_tokens": 3670757.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.938638305664062,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.8411,
|
|
"mean_token_accuracy": 0.14908881187438966,
|
|
"num_tokens": 3681286.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.86508994102478,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.8322,
|
|
"mean_token_accuracy": 0.15267545282840728,
|
|
"num_tokens": 3690733.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.98020076751709,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 6.0158,
|
|
"mean_token_accuracy": 0.14166081100702285,
|
|
"num_tokens": 3701263.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.95297589302063,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.9148,
|
|
"mean_token_accuracy": 0.14752905070781708,
|
|
"num_tokens": 3711578.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.993722248077392,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.9086,
|
|
"mean_token_accuracy": 0.14346786960959435,
|
|
"num_tokens": 3722140.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.804684638977051,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.6267,
|
|
"mean_token_accuracy": 0.1540088638663292,
|
|
"num_tokens": 3732019.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.840555429458618,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.8349,
|
|
"mean_token_accuracy": 0.14961637333035469,
|
|
"num_tokens": 3741912.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.951719141006469,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.14838206619024277,
|
|
"num_tokens": 3751365.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.806200456619263,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.8056,
|
|
"mean_token_accuracy": 0.14574428796768188,
|
|
"num_tokens": 3762160.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.896996021270752,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.8481,
|
|
"mean_token_accuracy": 0.14969197660684586,
|
|
"num_tokens": 3772608.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.856444787979126,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.14442452415823936,
|
|
"num_tokens": 3783344.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.943165874481201,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.14382675886154175,
|
|
"num_tokens": 3794103.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.821027517318726,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.82,
|
|
"mean_token_accuracy": 0.15115977749228476,
|
|
"num_tokens": 3804537.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.911714172363281,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.8419,
|
|
"mean_token_accuracy": 0.14932646602392197,
|
|
"num_tokens": 3814147.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.96919002532959,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.14370580911636352,
|
|
"num_tokens": 3823640.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.846808624267578,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.7682,
|
|
"mean_token_accuracy": 0.1589408814907074,
|
|
"num_tokens": 3833418.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.839104270935058,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.788,
|
|
"mean_token_accuracy": 0.15187744349241256,
|
|
"num_tokens": 3844998.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.784125089645386,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.7262,
|
|
"mean_token_accuracy": 0.15564873069524765,
|
|
"num_tokens": 3855549.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.813870859146118,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.8403,
|
|
"mean_token_accuracy": 0.15067547261714936,
|
|
"num_tokens": 3865287.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.872530603408814,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8987,
|
|
"mean_token_accuracy": 0.14980295076966285,
|
|
"num_tokens": 3875914.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.950064992904663,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.14314936101436615,
|
|
"num_tokens": 3885196.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.89986743927002,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.9203,
|
|
"mean_token_accuracy": 0.13913992792367935,
|
|
"num_tokens": 3895227.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.864428567886352,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.7911,
|
|
"mean_token_accuracy": 0.14839531779289244,
|
|
"num_tokens": 3905926.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.850302410125733,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.15205959975719452,
|
|
"num_tokens": 3916982.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.825182247161865,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.14805149286985397,
|
|
"num_tokens": 3926889.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.894171237945557,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.7834,
|
|
"mean_token_accuracy": 0.1433366745710373,
|
|
"num_tokens": 3936734.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.856796789169311,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.9236,
|
|
"mean_token_accuracy": 0.14193570539355277,
|
|
"num_tokens": 3948001.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.9134862422943115,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.14628526121377944,
|
|
"num_tokens": 3957433.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.915403604507446,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.888,
|
|
"mean_token_accuracy": 0.1475825399160385,
|
|
"num_tokens": 3967375.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.898691177368164,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.1420122355222702,
|
|
"num_tokens": 3977190.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.882886171340942,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.8512,
|
|
"mean_token_accuracy": 0.1531060181558132,
|
|
"num_tokens": 3987878.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.875026750564575,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.845,
|
|
"mean_token_accuracy": 0.14857483953237532,
|
|
"num_tokens": 3997166.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.897173833847046,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.8326,
|
|
"mean_token_accuracy": 0.14652054458856584,
|
|
"num_tokens": 4007711.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.9265892028808596,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.9763,
|
|
"mean_token_accuracy": 0.14133203625679017,
|
|
"num_tokens": 4018907.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.867978954315186,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.864,
|
|
"mean_token_accuracy": 0.1498611330986023,
|
|
"num_tokens": 4030006.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.989753198623657,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.9589,
|
|
"mean_token_accuracy": 0.1504744090139866,
|
|
"num_tokens": 4040609.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.826623916625977,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.7838,
|
|
"mean_token_accuracy": 0.14803807586431503,
|
|
"num_tokens": 4050392.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.823447513580322,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.1428491950035095,
|
|
"num_tokens": 4060156.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.940966033935547,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.7626,
|
|
"mean_token_accuracy": 0.15149431079626083,
|
|
"num_tokens": 4069670.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.776753520965576,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.7935,
|
|
"mean_token_accuracy": 0.1449214018881321,
|
|
"num_tokens": 4080738.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.7570106983184814,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.1536009728908539,
|
|
"num_tokens": 4091122.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.933842706680298,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.8733,
|
|
"mean_token_accuracy": 0.14600196480751038,
|
|
"num_tokens": 4101055.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.904268646240235,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.8308,
|
|
"mean_token_accuracy": 0.15079639106988907,
|
|
"num_tokens": 4110631.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.8311481952667235,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.14489958733320235,
|
|
"num_tokens": 4121120.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.924700593948364,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7857,
|
|
"mean_token_accuracy": 0.14904456734657287,
|
|
"num_tokens": 4132106.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.833415603637695,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.8293,
|
|
"mean_token_accuracy": 0.1477825567126274,
|
|
"num_tokens": 4142216.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.819109145985093,
|
|
"eval_loss": 6.050093173980713,
|
|
"eval_mean_token_accuracy": 0.14379453367965167,
|
|
"eval_num_tokens": 4142216.0,
|
|
"eval_runtime": 2.5968,
|
|
"eval_samples_per_second": 1322.778,
|
|
"eval_steps_per_second": 165.588,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.883230972290039,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.8701,
|
|
"mean_token_accuracy": 0.14971277937293054,
|
|
"num_tokens": 4152168.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.882227420806885,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.14443953037261964,
|
|
"num_tokens": 4162085.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.899635696411133,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.88,
|
|
"mean_token_accuracy": 0.14574271067976952,
|
|
"num_tokens": 4171168.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.8919424533844,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.1483355924487114,
|
|
"num_tokens": 4181804.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.960490989685058,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8872,
|
|
"mean_token_accuracy": 0.14367986619472503,
|
|
"num_tokens": 4192483.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.817529487609863,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.7738,
|
|
"mean_token_accuracy": 0.15300548374652861,
|
|
"num_tokens": 4202214.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.889630556106567,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.8652,
|
|
"mean_token_accuracy": 0.1525376059114933,
|
|
"num_tokens": 4212440.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.852598476409912,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.8168,
|
|
"mean_token_accuracy": 0.15018621683120728,
|
|
"num_tokens": 4222780.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.833838605880738,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.7536,
|
|
"mean_token_accuracy": 0.1566378265619278,
|
|
"num_tokens": 4232895.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.815893507003784,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.15336491912603378,
|
|
"num_tokens": 4244287.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.836065149307251,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.8025,
|
|
"mean_token_accuracy": 0.1468632087111473,
|
|
"num_tokens": 4254327.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.906033039093018,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.8389,
|
|
"mean_token_accuracy": 0.1460307225584984,
|
|
"num_tokens": 4264540.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.867164373397827,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.8125,
|
|
"mean_token_accuracy": 0.1475397951900959,
|
|
"num_tokens": 4274946.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.828709363937378,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.7574,
|
|
"mean_token_accuracy": 0.15103001445531844,
|
|
"num_tokens": 4284395.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.880394220352173,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.8516,
|
|
"mean_token_accuracy": 0.1469341918826103,
|
|
"num_tokens": 4294887.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.829086589813232,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.856,
|
|
"mean_token_accuracy": 0.1443422928452492,
|
|
"num_tokens": 4304794.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.8116350173950195,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.7663,
|
|
"mean_token_accuracy": 0.148785400390625,
|
|
"num_tokens": 4314810.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.871056842803955,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.8192,
|
|
"mean_token_accuracy": 0.14579248204827308,
|
|
"num_tokens": 4325444.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.896946716308594,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.8431,
|
|
"mean_token_accuracy": 0.1450091317296028,
|
|
"num_tokens": 4335620.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.843059062957764,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.8229,
|
|
"mean_token_accuracy": 0.15126920714974404,
|
|
"num_tokens": 4345006.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.874631357192993,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.7532,
|
|
"mean_token_accuracy": 0.14949243068695067,
|
|
"num_tokens": 4354805.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.850671720504761,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.1483994260430336,
|
|
"num_tokens": 4365881.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.93496618270874,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.14093863144516944,
|
|
"num_tokens": 4376368.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.822841024398803,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.8036,
|
|
"mean_token_accuracy": 0.15454284250736236,
|
|
"num_tokens": 4386695.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.945070791244507,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.14849908351898194,
|
|
"num_tokens": 4396837.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.75476770401001,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.7906,
|
|
"mean_token_accuracy": 0.1520374372601509,
|
|
"num_tokens": 4407310.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.8474101543426515,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.8292,
|
|
"mean_token_accuracy": 0.15293561220169066,
|
|
"num_tokens": 4417544.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.841001033782959,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.7706,
|
|
"mean_token_accuracy": 0.1506148397922516,
|
|
"num_tokens": 4428228.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.782240533828736,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7882,
|
|
"mean_token_accuracy": 0.1523190841078758,
|
|
"num_tokens": 4439088.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.873093318939209,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.8911,
|
|
"mean_token_accuracy": 0.14140778854489328,
|
|
"num_tokens": 4449715.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.786935186386108,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.7443,
|
|
"mean_token_accuracy": 0.15438926517963408,
|
|
"num_tokens": 4460233.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.798536014556885,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.6887,
|
|
"mean_token_accuracy": 0.15654485523700715,
|
|
"num_tokens": 4471401.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.7939526557922365,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.7489,
|
|
"mean_token_accuracy": 0.1589686781167984,
|
|
"num_tokens": 4481185.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.780501890182495,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.7891,
|
|
"mean_token_accuracy": 0.1477307014167309,
|
|
"num_tokens": 4492342.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.935231685638428,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8854,
|
|
"mean_token_accuracy": 0.1487235024571419,
|
|
"num_tokens": 4504277.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.8527364253997805,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.8389,
|
|
"mean_token_accuracy": 0.1489283800125122,
|
|
"num_tokens": 4514178.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.838755083084107,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.7981,
|
|
"mean_token_accuracy": 0.15372906774282455,
|
|
"num_tokens": 4524559.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.845714664459228,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.7546,
|
|
"mean_token_accuracy": 0.15459949672222137,
|
|
"num_tokens": 4535168.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.879129028320312,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.14930276274681092,
|
|
"num_tokens": 4546105.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.856486082077026,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.77,
|
|
"mean_token_accuracy": 0.15297604352235794,
|
|
"num_tokens": 4556498.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.700125482347277,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.5762,
|
|
"mean_token_accuracy": 0.14985370801554787,
|
|
"num_tokens": 4565029.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.850874900817871,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.5442,
|
|
"mean_token_accuracy": 0.154378442466259,
|
|
"num_tokens": 4575232.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.784826231002808,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.5174,
|
|
"mean_token_accuracy": 0.15894271433353424,
|
|
"num_tokens": 4585818.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.703923082351684,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.481,
|
|
"mean_token_accuracy": 0.16246529668569565,
|
|
"num_tokens": 4595823.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.780557489395141,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.16049877256155015,
|
|
"num_tokens": 4606537.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.799535465240479,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.15943138301372528,
|
|
"num_tokens": 4616096.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.776993227005005,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.5149,
|
|
"mean_token_accuracy": 0.15986263900995254,
|
|
"num_tokens": 4626287.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.737018346786499,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.5113,
|
|
"mean_token_accuracy": 0.15839564353227614,
|
|
"num_tokens": 4636898.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.675017547607422,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.4703,
|
|
"mean_token_accuracy": 0.16265783458948135,
|
|
"num_tokens": 4646263.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.818978929519654,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.1600583478808403,
|
|
"num_tokens": 4656463.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.7211040496826175,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.5389,
|
|
"mean_token_accuracy": 0.16089007407426834,
|
|
"num_tokens": 4666824.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.666584920883179,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.4265,
|
|
"mean_token_accuracy": 0.16442574113607406,
|
|
"num_tokens": 4676793.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.815246295928955,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.15934282541275024,
|
|
"num_tokens": 4687124.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.7452959537506105,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.7288,
|
|
"mean_token_accuracy": 0.15075535774230958,
|
|
"num_tokens": 4698074.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.82332558631897,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.5919,
|
|
"mean_token_accuracy": 0.15656624138355255,
|
|
"num_tokens": 4709177.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.848820304870605,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.5109,
|
|
"mean_token_accuracy": 0.15775267779827118,
|
|
"num_tokens": 4719649.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.709406423568725,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.5799,
|
|
"mean_token_accuracy": 0.15765960812568663,
|
|
"num_tokens": 4731685.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.7174700736999515,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.15760382562875747,
|
|
"num_tokens": 4742854.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.7340339660644535,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.5131,
|
|
"mean_token_accuracy": 0.15671638399362564,
|
|
"num_tokens": 4754088.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.714247369766236,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.5341,
|
|
"mean_token_accuracy": 0.15981189608573915,
|
|
"num_tokens": 4764090.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.711434459686279,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.5242,
|
|
"mean_token_accuracy": 0.16334791630506515,
|
|
"num_tokens": 4773704.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.770476484298706,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.554,
|
|
"mean_token_accuracy": 0.1621487781405449,
|
|
"num_tokens": 4783678.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.663452053070069,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.16795776933431625,
|
|
"num_tokens": 4794183.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.65028862953186,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.4312,
|
|
"mean_token_accuracy": 0.16061999499797822,
|
|
"num_tokens": 4803670.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.70339150428772,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.4528,
|
|
"mean_token_accuracy": 0.16460076123476028,
|
|
"num_tokens": 4813611.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.655531644821167,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.5207,
|
|
"mean_token_accuracy": 0.15569410920143129,
|
|
"num_tokens": 4824356.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.789508867263794,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.15906214267015456,
|
|
"num_tokens": 4834410.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.648446416854858,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.4117,
|
|
"mean_token_accuracy": 0.1643962562084198,
|
|
"num_tokens": 4844975.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.662750911712647,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.5309,
|
|
"mean_token_accuracy": 0.16364205926656722,
|
|
"num_tokens": 4856419.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.66937484741211,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.4444,
|
|
"mean_token_accuracy": 0.165959732234478,
|
|
"num_tokens": 4867115.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.703761243820191,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.5108,
|
|
"mean_token_accuracy": 0.16079539358615874,
|
|
"num_tokens": 4876675.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.720136308670044,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.5404,
|
|
"mean_token_accuracy": 0.1617947369813919,
|
|
"num_tokens": 4887868.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.6826314449310305,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.5123,
|
|
"mean_token_accuracy": 0.16172491312026976,
|
|
"num_tokens": 4896878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.664020633697509,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4854,
|
|
"mean_token_accuracy": 0.16069988757371903,
|
|
"num_tokens": 4906683.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.680099058151245,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.4707,
|
|
"mean_token_accuracy": 0.16708637923002242,
|
|
"num_tokens": 4917184.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.683479261398316,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.5037,
|
|
"mean_token_accuracy": 0.15707164853811265,
|
|
"num_tokens": 4927337.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.729428672790528,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.5124,
|
|
"mean_token_accuracy": 0.15443907007575036,
|
|
"num_tokens": 4937661.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.729527378082276,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.15152359902858734,
|
|
"num_tokens": 4948068.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.660752439498902,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4923,
|
|
"mean_token_accuracy": 0.16463814526796341,
|
|
"num_tokens": 4958703.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.6649699211120605,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.5129,
|
|
"mean_token_accuracy": 0.1626749038696289,
|
|
"num_tokens": 4969190.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.644364356994629,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.4013,
|
|
"mean_token_accuracy": 0.16723495423793794,
|
|
"num_tokens": 4978663.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.743815612792969,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.5922,
|
|
"mean_token_accuracy": 0.16085838675498962,
|
|
"num_tokens": 4989839.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.603341817855835,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.16587418615818023,
|
|
"num_tokens": 5000030.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.701595258712769,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.5925,
|
|
"mean_token_accuracy": 0.16056383699178695,
|
|
"num_tokens": 5010229.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.726330900192261,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.5118,
|
|
"mean_token_accuracy": 0.1618313744664192,
|
|
"num_tokens": 5021000.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.677479362487793,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.4199,
|
|
"mean_token_accuracy": 0.16339841037988662,
|
|
"num_tokens": 5031319.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.613199663162232,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.4571,
|
|
"mean_token_accuracy": 0.16850902140140533,
|
|
"num_tokens": 5041892.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.629352140426636,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.5338,
|
|
"mean_token_accuracy": 0.16332726627588273,
|
|
"num_tokens": 5052363.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.762828731536866,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.5622,
|
|
"mean_token_accuracy": 0.15506864935159684,
|
|
"num_tokens": 5063342.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.688890409469605,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.4758,
|
|
"mean_token_accuracy": 0.15885344594717027,
|
|
"num_tokens": 5073393.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.57670488357544,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.4403,
|
|
"mean_token_accuracy": 0.1676219493150711,
|
|
"num_tokens": 5082417.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.78955774307251,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.6069,
|
|
"mean_token_accuracy": 0.15390974432229995,
|
|
"num_tokens": 5093383.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.580852699279785,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.3906,
|
|
"mean_token_accuracy": 0.16938741654157638,
|
|
"num_tokens": 5104346.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.638398218154907,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.561,
|
|
"mean_token_accuracy": 0.15911775380373,
|
|
"num_tokens": 5114375.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.732828569412232,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.5355,
|
|
"mean_token_accuracy": 0.1595481425523758,
|
|
"num_tokens": 5125035.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.6798309803009035,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.5463,
|
|
"mean_token_accuracy": 0.152773617208004,
|
|
"num_tokens": 5135869.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.720887184143066,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.5134,
|
|
"mean_token_accuracy": 0.15898497104644777,
|
|
"num_tokens": 5146615.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.560636568069458,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.4127,
|
|
"mean_token_accuracy": 0.16486843973398208,
|
|
"num_tokens": 5157490.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.648875522613525,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.55,
|
|
"mean_token_accuracy": 0.15591932386159896,
|
|
"num_tokens": 5168693.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.7419291019439695,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.5106,
|
|
"mean_token_accuracy": 0.15850055515766143,
|
|
"num_tokens": 5179232.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.491250712372536,
|
|
"eval_loss": 5.986842632293701,
|
|
"eval_mean_token_accuracy": 0.14950340444785218,
|
|
"eval_num_tokens": 5179232.0,
|
|
"eval_runtime": 2.5975,
|
|
"eval_samples_per_second": 1322.426,
|
|
"eval_steps_per_second": 165.544,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.627499151229858,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.5545,
|
|
"mean_token_accuracy": 0.1652067631483078,
|
|
"num_tokens": 5188613.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.633312320709228,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.4399,
|
|
"mean_token_accuracy": 0.1668851926922798,
|
|
"num_tokens": 5198441.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.610636854171753,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.16499845832586288,
|
|
"num_tokens": 5208466.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.565611696243286,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.4865,
|
|
"mean_token_accuracy": 0.16280112117528917,
|
|
"num_tokens": 5218847.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.737601947784424,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.15756618082523347,
|
|
"num_tokens": 5228870.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.713624429702759,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.4729,
|
|
"mean_token_accuracy": 0.16696099787950516,
|
|
"num_tokens": 5238186.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.617369508743286,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.5339,
|
|
"mean_token_accuracy": 0.16417384147644043,
|
|
"num_tokens": 5247478.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.709841299057007,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.4687,
|
|
"mean_token_accuracy": 0.1602501317858696,
|
|
"num_tokens": 5258519.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.698099279403687,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.5135,
|
|
"mean_token_accuracy": 0.1668350026011467,
|
|
"num_tokens": 5268218.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.506252813339233,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.4552,
|
|
"mean_token_accuracy": 0.1638593629002571,
|
|
"num_tokens": 5277926.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.658507204055786,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4999,
|
|
"mean_token_accuracy": 0.16297332048416138,
|
|
"num_tokens": 5287735.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.676133441925049,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.4182,
|
|
"mean_token_accuracy": 0.17203330993652344,
|
|
"num_tokens": 5297120.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.489224767684936,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.4428,
|
|
"mean_token_accuracy": 0.16497790664434434,
|
|
"num_tokens": 5307448.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.675883531570435,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.4505,
|
|
"mean_token_accuracy": 0.16972960382699967,
|
|
"num_tokens": 5317914.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.675877332687378,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.4562,
|
|
"mean_token_accuracy": 0.16813937127590178,
|
|
"num_tokens": 5328471.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.571584844589234,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.462,
|
|
"mean_token_accuracy": 0.1582278087735176,
|
|
"num_tokens": 5339118.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.678888940811158,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.568,
|
|
"mean_token_accuracy": 0.15406795740127563,
|
|
"num_tokens": 5350183.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.663352632522583,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.4974,
|
|
"mean_token_accuracy": 0.15896444842219354,
|
|
"num_tokens": 5361317.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.649419784545898,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.4983,
|
|
"mean_token_accuracy": 0.16494401395320893,
|
|
"num_tokens": 5371338.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.732968187332153,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.6042,
|
|
"mean_token_accuracy": 0.15870766192674637,
|
|
"num_tokens": 5382043.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.698143482208252,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.5121,
|
|
"mean_token_accuracy": 0.16373514086008073,
|
|
"num_tokens": 5392547.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.527518653869629,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.3983,
|
|
"mean_token_accuracy": 0.17403582781553267,
|
|
"num_tokens": 5402455.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.584602308273316,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.5633,
|
|
"mean_token_accuracy": 0.15951344221830369,
|
|
"num_tokens": 5413925.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.713425874710083,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.5907,
|
|
"mean_token_accuracy": 0.1547597050666809,
|
|
"num_tokens": 5425344.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.61894736289978,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.16069943457841873,
|
|
"num_tokens": 5435197.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.728053712844849,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.5248,
|
|
"mean_token_accuracy": 0.16462774872779845,
|
|
"num_tokens": 5445683.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.610408639907837,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.3851,
|
|
"mean_token_accuracy": 0.1635379433631897,
|
|
"num_tokens": 5456846.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.581979417800904,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.4559,
|
|
"mean_token_accuracy": 0.1690095007419586,
|
|
"num_tokens": 5467701.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.623260831832885,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.5023,
|
|
"mean_token_accuracy": 0.15982526838779448,
|
|
"num_tokens": 5478246.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.644440841674805,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.16287316530942916,
|
|
"num_tokens": 5488107.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.7388874053955075,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.4915,
|
|
"mean_token_accuracy": 0.16001203134655953,
|
|
"num_tokens": 5498521.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.569246292114258,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.5285,
|
|
"mean_token_accuracy": 0.1563582494854927,
|
|
"num_tokens": 5509450.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.6742730140686035,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.16480647027492523,
|
|
"num_tokens": 5519662.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.586445379257202,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.16713995933532716,
|
|
"num_tokens": 5529499.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.622071361541748,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.15462029725313187,
|
|
"num_tokens": 5540938.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.657432556152344,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.16584471315145494,
|
|
"num_tokens": 5552120.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.603826475143433,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.4356,
|
|
"mean_token_accuracy": 0.16504934132099153,
|
|
"num_tokens": 5561590.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.567124700546264,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.4165,
|
|
"mean_token_accuracy": 0.1707810118794441,
|
|
"num_tokens": 5571682.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.615508699417115,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.4673,
|
|
"mean_token_accuracy": 0.1730094686150551,
|
|
"num_tokens": 5581299.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.545977163314819,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.4154,
|
|
"mean_token_accuracy": 0.16748601645231248,
|
|
"num_tokens": 5590766.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.622316646575928,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.5316,
|
|
"mean_token_accuracy": 0.16143546551465987,
|
|
"num_tokens": 5600898.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.66291880607605,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.5158,
|
|
"mean_token_accuracy": 0.16342089623212813,
|
|
"num_tokens": 5612052.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.660036611557007,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.5874,
|
|
"mean_token_accuracy": 0.16431321948766708,
|
|
"num_tokens": 5622847.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.567380332946778,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.17110821902751921,
|
|
"num_tokens": 5633134.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.652763843536377,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.5767,
|
|
"mean_token_accuracy": 0.16604768484830856,
|
|
"num_tokens": 5644223.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.593555927276611,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.5115,
|
|
"mean_token_accuracy": 0.16058846265077592,
|
|
"num_tokens": 5653620.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.677840805053711,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.4237,
|
|
"mean_token_accuracy": 0.16867171376943588,
|
|
"num_tokens": 5663800.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.608797216415406,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.16562926769256592,
|
|
"num_tokens": 5674606.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.664605760574341,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.5954,
|
|
"mean_token_accuracy": 0.16185092478990554,
|
|
"num_tokens": 5684975.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.600146913528443,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.435,
|
|
"mean_token_accuracy": 0.16718858629465103,
|
|
"num_tokens": 5695336.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.6710587501525875,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.4596,
|
|
"mean_token_accuracy": 0.16664838194847106,
|
|
"num_tokens": 5706027.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.599521589279175,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.1608511984348297,
|
|
"num_tokens": 5716026.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.6033525466918945,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.499,
|
|
"mean_token_accuracy": 0.16632484197616576,
|
|
"num_tokens": 5726055.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.6622168064117435,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.542,
|
|
"mean_token_accuracy": 0.15696141570806504,
|
|
"num_tokens": 5736562.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.73774094581604,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.5769,
|
|
"mean_token_accuracy": 0.15729292780160903,
|
|
"num_tokens": 5747297.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.619591951370239,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.5425,
|
|
"mean_token_accuracy": 0.16078144162893296,
|
|
"num_tokens": 5756801.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.670393371582032,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.5367,
|
|
"mean_token_accuracy": 0.1635377436876297,
|
|
"num_tokens": 5767493.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.616514539718628,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.1742817223072052,
|
|
"num_tokens": 5777539.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.591523456573486,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.167883762717247,
|
|
"num_tokens": 5787202.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.6892999649047855,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.5049,
|
|
"mean_token_accuracy": 0.1640034094452858,
|
|
"num_tokens": 5798023.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.523151063919068,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.4072,
|
|
"mean_token_accuracy": 0.16618600934743882,
|
|
"num_tokens": 5807711.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.562091875076294,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.412,
|
|
"mean_token_accuracy": 0.16975668370723723,
|
|
"num_tokens": 5817053.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.569506883621216,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.5157,
|
|
"mean_token_accuracy": 0.16288599371910095,
|
|
"num_tokens": 5827792.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.600739526748657,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.4074,
|
|
"mean_token_accuracy": 0.16991793662309645,
|
|
"num_tokens": 5837702.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.692067098617554,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.5546,
|
|
"mean_token_accuracy": 0.16360698789358138,
|
|
"num_tokens": 5847973.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.637374544143677,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.5499,
|
|
"mean_token_accuracy": 0.15725272595882417,
|
|
"num_tokens": 5857591.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.565674448013306,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.5059,
|
|
"mean_token_accuracy": 0.1634950265288353,
|
|
"num_tokens": 5868616.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.650606632232666,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.4348,
|
|
"mean_token_accuracy": 0.16790513396263124,
|
|
"num_tokens": 5878526.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.590163040161133,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.1749364659190178,
|
|
"num_tokens": 5887844.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.644021272659302,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.5451,
|
|
"mean_token_accuracy": 0.16367463767528534,
|
|
"num_tokens": 5898523.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.603885507583618,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.5536,
|
|
"mean_token_accuracy": 0.15856787636876107,
|
|
"num_tokens": 5909671.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.59442548751831,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.4891,
|
|
"mean_token_accuracy": 0.15759917497634887,
|
|
"num_tokens": 5920451.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.675187778472901,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.4765,
|
|
"mean_token_accuracy": 0.16859863847494125,
|
|
"num_tokens": 5931218.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.48574743270874,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.4751,
|
|
"mean_token_accuracy": 0.16344277709722518,
|
|
"num_tokens": 5941380.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.614834976196289,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.16504250317811966,
|
|
"num_tokens": 5951989.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.634076070785523,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.5061,
|
|
"mean_token_accuracy": 0.1620146870613098,
|
|
"num_tokens": 5961982.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.624727010726929,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.5309,
|
|
"mean_token_accuracy": 0.16035393476486207,
|
|
"num_tokens": 5972184.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.671657609939575,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.4491,
|
|
"mean_token_accuracy": 0.168126380443573,
|
|
"num_tokens": 5981912.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.61940746307373,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.16417238712310792,
|
|
"num_tokens": 5992460.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.6031458377838135,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.5058,
|
|
"mean_token_accuracy": 0.16285987943410873,
|
|
"num_tokens": 6002170.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.573065280914307,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.4317,
|
|
"mean_token_accuracy": 0.1615881845355034,
|
|
"num_tokens": 6012240.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.630006313323975,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.4057,
|
|
"mean_token_accuracy": 0.16957840621471404,
|
|
"num_tokens": 6022289.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.642009019851685,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.524,
|
|
"mean_token_accuracy": 0.15756816565990447,
|
|
"num_tokens": 6032268.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.702442169189453,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.5905,
|
|
"mean_token_accuracy": 0.15488297939300538,
|
|
"num_tokens": 6042574.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.769234132766724,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.5579,
|
|
"mean_token_accuracy": 0.1582654058933258,
|
|
"num_tokens": 6052508.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.549481201171875,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.1712830290198326,
|
|
"num_tokens": 6062541.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.540633487701416,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.5691,
|
|
"mean_token_accuracy": 0.15504138320684432,
|
|
"num_tokens": 6072532.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.653427791595459,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 2.703125,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.16891696751117707,
|
|
"num_tokens": 6082518.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.657943344116211,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.4468,
|
|
"mean_token_accuracy": 0.164460127055645,
|
|
"num_tokens": 6092356.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.634619855880738,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.5382,
|
|
"mean_token_accuracy": 0.165224651992321,
|
|
"num_tokens": 6103450.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.653062343597412,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.17163696736097336,
|
|
"num_tokens": 6113708.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.558749485015869,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.5134,
|
|
"mean_token_accuracy": 0.16230242252349852,
|
|
"num_tokens": 6124508.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.692312717437744,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.4872,
|
|
"mean_token_accuracy": 0.17191842198371887,
|
|
"num_tokens": 6135516.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.566852283477783,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.412,
|
|
"mean_token_accuracy": 0.16392800360918044,
|
|
"num_tokens": 6146206.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.518176746368408,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.383,
|
|
"mean_token_accuracy": 0.17127594947814942,
|
|
"num_tokens": 6156244.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.447549772262573,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.4112,
|
|
"mean_token_accuracy": 0.16063664257526397,
|
|
"num_tokens": 6167091.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.6188061237335205,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.16669660359621047,
|
|
"num_tokens": 6177993.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.576446056365967,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.5158,
|
|
"mean_token_accuracy": 0.16666642427444459,
|
|
"num_tokens": 6187963.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.595619201660156,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.4884,
|
|
"mean_token_accuracy": 0.1625219315290451,
|
|
"num_tokens": 6198442.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.610590934753418,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.4934,
|
|
"mean_token_accuracy": 0.1641159787774086,
|
|
"num_tokens": 6209332.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.522899136432382,
|
|
"eval_loss": 5.916890621185303,
|
|
"eval_mean_token_accuracy": 0.15391022185253542,
|
|
"eval_num_tokens": 6209332.0,
|
|
"eval_runtime": 2.5974,
|
|
"eval_samples_per_second": 1322.454,
|
|
"eval_steps_per_second": 165.547,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.542118501663208,
|
|
"epoch": 2.729668332576102,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004568624876891796,
|
|
"loss": 5.4611,
|
|
"mean_token_accuracy": 0.16434952467679978,
|
|
"num_tokens": 6219467.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.56941180229187,
|
|
"epoch": 2.7342117219445705,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00045665416380864587,
|
|
"loss": 5.4806,
|
|
"mean_token_accuracy": 0.16400491297245026,
|
|
"num_tokens": 6229277.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.6014204025268555,
|
|
"epoch": 2.7387551113130395,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004564453917145206,
|
|
"loss": 5.4096,
|
|
"mean_token_accuracy": 0.17019634395837785,
|
|
"num_tokens": 6239191.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.62946047782898,
|
|
"epoch": 2.7432985006815085,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004562361719219283,
|
|
"loss": 5.5286,
|
|
"mean_token_accuracy": 0.15971722453832626,
|
|
"num_tokens": 6249539.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.610321807861328,
|
|
"epoch": 2.747841890049977,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00045602650494709786,
|
|
"loss": 5.5419,
|
|
"mean_token_accuracy": 0.15591299384832383,
|
|
"num_tokens": 6260896.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.642046976089477,
|
|
"epoch": 2.752385279418446,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045581639130736173,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.1603671833872795,
|
|
"num_tokens": 6271828.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.7217706680297855,
|
|
"epoch": 2.756928668786915,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004556058315211546,
|
|
"loss": 5.5258,
|
|
"mean_token_accuracy": 0.1675170913338661,
|
|
"num_tokens": 6282698.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.603877973556519,
|
|
"epoch": 2.761472058155384,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045539482610801156,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.16894223988056184,
|
|
"num_tokens": 6291905.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.514861536026001,
|
|
"epoch": 2.766015447523853,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004551833755885677,
|
|
"loss": 5.4646,
|
|
"mean_token_accuracy": 0.16589600890874862,
|
|
"num_tokens": 6301579.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.605765104293823,
|
|
"epoch": 2.7705588368923215,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004549714804845561,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.163409923017025,
|
|
"num_tokens": 6311209.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.548905897140503,
|
|
"epoch": 2.7751022262607905,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0004547591413188068,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.16202747374773024,
|
|
"num_tokens": 6321601.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.635558700561523,
|
|
"epoch": 2.7796456156292595,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004545463586152456,
|
|
"loss": 5.5434,
|
|
"mean_token_accuracy": 0.1665214255452156,
|
|
"num_tokens": 6331070.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.641869592666626,
|
|
"epoch": 2.784189004997728,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004543331328988927,
|
|
"loss": 5.5106,
|
|
"mean_token_accuracy": 0.16080092936754226,
|
|
"num_tokens": 6341772.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.501345205307007,
|
|
"epoch": 2.788732394366197,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004541194646958614,
|
|
"loss": 5.4593,
|
|
"mean_token_accuracy": 0.16531397253274918,
|
|
"num_tokens": 6352237.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.66186408996582,
|
|
"epoch": 2.793275783734666,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045390535453335666,
|
|
"loss": 5.5752,
|
|
"mean_token_accuracy": 0.1561602458357811,
|
|
"num_tokens": 6363645.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.593492650985718,
|
|
"epoch": 2.797819173103135,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045369080293967414,
|
|
"loss": 5.4502,
|
|
"mean_token_accuracy": 0.16789142787456512,
|
|
"num_tokens": 6373559.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.634049940109253,
|
|
"epoch": 2.802362562471604,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00045347581044419846,
|
|
"loss": 5.4521,
|
|
"mean_token_accuracy": 0.16563277840614318,
|
|
"num_tokens": 6384197.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.543875408172608,
|
|
"epoch": 2.806905951840073,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045326037757740216,
|
|
"loss": 5.3581,
|
|
"mean_token_accuracy": 0.16752002090215684,
|
|
"num_tokens": 6394834.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.521064758300781,
|
|
"epoch": 2.8114493412085415,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004530445048708447,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.16238887161016463,
|
|
"num_tokens": 6405515.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.627286958694458,
|
|
"epoch": 2.8159927305770105,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045282819285717024,
|
|
"loss": 5.4825,
|
|
"mean_token_accuracy": 0.1663505971431732,
|
|
"num_tokens": 6415885.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.531750583648682,
|
|
"epoch": 2.8205361199454795,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045261144207010736,
|
|
"loss": 5.4094,
|
|
"mean_token_accuracy": 0.17296303808689117,
|
|
"num_tokens": 6426053.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.590877294540405,
|
|
"epoch": 2.825079509313948,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045239425304446684,
|
|
"loss": 5.5346,
|
|
"mean_token_accuracy": 0.15836047381162643,
|
|
"num_tokens": 6436024.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.668206977844238,
|
|
"epoch": 2.829622898682417,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004521766263161413,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.16192781776189805,
|
|
"num_tokens": 6446847.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.558481979370117,
|
|
"epoch": 2.834166288050886,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000451958562422103,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.16717706322669984,
|
|
"num_tokens": 6456901.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.591206884384155,
|
|
"epoch": 2.838709677419355,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00045174006190040267,
|
|
"loss": 5.4445,
|
|
"mean_token_accuracy": 0.16293987035751342,
|
|
"num_tokens": 6467308.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.593704891204834,
|
|
"epoch": 2.843253066787824,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000451521125290169,
|
|
"loss": 5.4211,
|
|
"mean_token_accuracy": 0.17131438106298447,
|
|
"num_tokens": 6477071.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.549534797668457,
|
|
"epoch": 2.8477964561562925,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000451301753131606,
|
|
"loss": 5.4829,
|
|
"mean_token_accuracy": 0.16538456529378892,
|
|
"num_tokens": 6488300.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.593517780303955,
|
|
"epoch": 2.8523398455247615,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00045108194596599286,
|
|
"loss": 5.5347,
|
|
"mean_token_accuracy": 0.1554817885160446,
|
|
"num_tokens": 6499515.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.576261425018311,
|
|
"epoch": 2.8568832348932305,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004508617043356818,
|
|
"loss": 5.4665,
|
|
"mean_token_accuracy": 0.16541536748409272,
|
|
"num_tokens": 6509167.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.620144462585449,
|
|
"epoch": 2.861426624261699,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004506410287840972,
|
|
"loss": 5.4938,
|
|
"mean_token_accuracy": 0.1609639495611191,
|
|
"num_tokens": 6520899.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.574986457824707,
|
|
"epoch": 2.865970013630168,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000450419919855734,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.1607181578874588,
|
|
"num_tokens": 6531293.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.5995519161224365,
|
|
"epoch": 2.870513402998637,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0004501983780961566,
|
|
"loss": 5.4723,
|
|
"mean_token_accuracy": 0.16596077978610993,
|
|
"num_tokens": 6541039.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.549083566665649,
|
|
"epoch": 2.875056792367106,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004499764040519972,
|
|
"loss": 5.4627,
|
|
"mean_token_accuracy": 0.16740722060203553,
|
|
"num_tokens": 6551948.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.598315238952637,
|
|
"epoch": 2.879600181735575,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044975399827095467,
|
|
"loss": 5.3923,
|
|
"mean_token_accuracy": 0.17182673662900924,
|
|
"num_tokens": 6562158.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.658646774291992,
|
|
"epoch": 2.8841435711040435,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044953116130179313,
|
|
"loss": 5.4917,
|
|
"mean_token_accuracy": 0.15569328814744948,
|
|
"num_tokens": 6572605.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.440945434570312,
|
|
"epoch": 2.8886869604725125,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004493078936943408,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.16961103230714797,
|
|
"num_tokens": 6583247.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.695300912857055,
|
|
"epoch": 2.8932303498409815,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004490841959994882,
|
|
"loss": 5.5686,
|
|
"mean_token_accuracy": 0.1563364863395691,
|
|
"num_tokens": 6593371.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.651923990249633,
|
|
"epoch": 2.89777373920945,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044886006876918716,
|
|
"loss": 5.3677,
|
|
"mean_token_accuracy": 0.16403858959674836,
|
|
"num_tokens": 6603777.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.640544652938843,
|
|
"epoch": 2.902317128577919,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004486355125564493,
|
|
"loss": 5.617,
|
|
"mean_token_accuracy": 0.1627153292298317,
|
|
"num_tokens": 6615654.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.612918472290039,
|
|
"epoch": 2.906860517946388,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004484105279153449,
|
|
"loss": 5.481,
|
|
"mean_token_accuracy": 0.1598746582865715,
|
|
"num_tokens": 6626079.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.603912115097046,
|
|
"epoch": 2.911403907314857,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000448185115401001,
|
|
"loss": 5.3813,
|
|
"mean_token_accuracy": 0.17635067105293273,
|
|
"num_tokens": 6636047.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.550787734985351,
|
|
"epoch": 2.915947296683326,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004479592755696008,
|
|
"loss": 5.5585,
|
|
"mean_token_accuracy": 0.1557467296719551,
|
|
"num_tokens": 6646670.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.618472671508789,
|
|
"epoch": 2.9204906860517945,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044773300897838137,
|
|
"loss": 5.56,
|
|
"mean_token_accuracy": 0.1680152416229248,
|
|
"num_tokens": 6658027.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.682929754257202,
|
|
"epoch": 2.9250340754202635,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004475063161856332,
|
|
"loss": 5.4835,
|
|
"mean_token_accuracy": 0.16703135073184966,
|
|
"num_tokens": 6668446.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.5976167678833,
|
|
"epoch": 2.9295774647887325,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004472791977506983,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.16511500924825667,
|
|
"num_tokens": 6679038.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.570468425750732,
|
|
"epoch": 2.934120854157201,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044705165423396855,
|
|
"loss": 5.4106,
|
|
"mean_token_accuracy": 0.17460913509130477,
|
|
"num_tokens": 6689207.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.579809331893921,
|
|
"epoch": 2.93866424352567,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00044682368619688523,
|
|
"loss": 5.4336,
|
|
"mean_token_accuracy": 0.16403188556432724,
|
|
"num_tokens": 6699264.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.5837079048156735,
|
|
"epoch": 2.943207632894139,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004465952942019367,
|
|
"loss": 5.4814,
|
|
"mean_token_accuracy": 0.16315781474113464,
|
|
"num_tokens": 6709309.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.583945989608765,
|
|
"epoch": 2.947751022262608,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044636647881265745,
|
|
"loss": 5.4598,
|
|
"mean_token_accuracy": 0.16575903296470643,
|
|
"num_tokens": 6719302.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.6043600082397464,
|
|
"epoch": 2.952294411631077,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004461372405936269,
|
|
"loss": 5.4156,
|
|
"mean_token_accuracy": 0.16708488315343856,
|
|
"num_tokens": 6730668.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.567113971710205,
|
|
"epoch": 2.9568378009995455,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004459075801104673,
|
|
"loss": 5.4059,
|
|
"mean_token_accuracy": 0.1614591509103775,
|
|
"num_tokens": 6740140.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.56418924331665,
|
|
"epoch": 2.9613811903680145,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044567749792984325,
|
|
"loss": 5.4405,
|
|
"mean_token_accuracy": 0.16615875363349913,
|
|
"num_tokens": 6751864.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.595006847381592,
|
|
"epoch": 2.9659245797364835,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.00044544699461945966,
|
|
"loss": 5.5575,
|
|
"mean_token_accuracy": 0.16195874959230422,
|
|
"num_tokens": 6762375.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.631557178497315,
|
|
"epoch": 2.970467969104952,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00044521607074806037,
|
|
"loss": 5.481,
|
|
"mean_token_accuracy": 0.16731629520654678,
|
|
"num_tokens": 6772181.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.606076002120972,
|
|
"epoch": 2.975011358473421,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004449847268854272,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.16299747973680495,
|
|
"num_tokens": 6782402.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.538914012908935,
|
|
"epoch": 2.97955474784189,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004447529636023782,
|
|
"loss": 5.4477,
|
|
"mean_token_accuracy": 0.16805306524038316,
|
|
"num_tokens": 6792671.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.545539808273316,
|
|
"epoch": 2.984098137210359,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00044452078147076614,
|
|
"loss": 5.3382,
|
|
"mean_token_accuracy": 0.17342765033245086,
|
|
"num_tokens": 6802111.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.557024621963501,
|
|
"epoch": 2.988641526578828,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004442881810634774,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.1697560280561447,
|
|
"num_tokens": 6812651.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.606369972229004,
|
|
"epoch": 2.9931849159472965,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004440551629544303,
|
|
"loss": 5.4691,
|
|
"mean_token_accuracy": 0.1571819603443146,
|
|
"num_tokens": 6823208.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.634217262268066,
|
|
"epoch": 2.9977283053157655,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00044382172771857383,
|
|
"loss": 5.4588,
|
|
"mean_token_accuracy": 0.16337359845638275,
|
|
"num_tokens": 6833412.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.580940617455377,
|
|
"epoch": 3.0018173557473875,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004435878759318864,
|
|
"loss": 5.3437,
|
|
"mean_token_accuracy": 0.17550832198725808,
|
|
"num_tokens": 6841977.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.554789781570435,
|
|
"epoch": 3.0063607451158565,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004433536081713738,
|
|
"loss": 5.1528,
|
|
"mean_token_accuracy": 0.17688107937574388,
|
|
"num_tokens": 6852571.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.574832582473755,
|
|
"epoch": 3.0109041344843255,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004431189250150686,
|
|
"loss": 5.2011,
|
|
"mean_token_accuracy": 0.1709735944867134,
|
|
"num_tokens": 6861451.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.565835571289062,
|
|
"epoch": 3.015447523852794,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00044288382704202816,
|
|
"loss": 5.1644,
|
|
"mean_token_accuracy": 0.17401813119649887,
|
|
"num_tokens": 6872134.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.51848726272583,
|
|
"epoch": 3.019990913221263,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044264831483233335,
|
|
"loss": 5.2244,
|
|
"mean_token_accuracy": 0.1795010522007942,
|
|
"num_tokens": 6882477.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.538839435577392,
|
|
"epoch": 3.024534302589732,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00044241238896708716,
|
|
"loss": 5.159,
|
|
"mean_token_accuracy": 0.1787806823849678,
|
|
"num_tokens": 6892233.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.501246976852417,
|
|
"epoch": 3.029077691958201,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0004421760500284131,
|
|
"loss": 5.1321,
|
|
"mean_token_accuracy": 0.17679936289787293,
|
|
"num_tokens": 6902090.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.521198511123657,
|
|
"epoch": 3.0336210813266695,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004419392985994541,
|
|
"loss": 5.23,
|
|
"mean_token_accuracy": 0.17728585004806519,
|
|
"num_tokens": 6912295.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.693401622772217,
|
|
"epoch": 3.0381644706951385,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044170213526437076,
|
|
"loss": 5.2554,
|
|
"mean_token_accuracy": 0.1622602179646492,
|
|
"num_tokens": 6923064.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.577483129501343,
|
|
"epoch": 3.0427078600636075,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044146456060833994,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.17781320512294768,
|
|
"num_tokens": 6932825.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.401189756393433,
|
|
"epoch": 3.0472512494320765,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00044122657521755373,
|
|
"loss": 5.0949,
|
|
"mean_token_accuracy": 0.18163834810256957,
|
|
"num_tokens": 6943188.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.481717348098755,
|
|
"epoch": 3.051794638800545,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00044098817967921716,
|
|
"loss": 5.1947,
|
|
"mean_token_accuracy": 0.17781210392713548,
|
|
"num_tokens": 6953864.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.508717584609985,
|
|
"epoch": 3.056338028169014,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044074937458154763,
|
|
"loss": 5.1865,
|
|
"mean_token_accuracy": 0.17617179155349733,
|
|
"num_tokens": 6963745.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.519685840606689,
|
|
"epoch": 3.060881417537483,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044051016051377303,
|
|
"loss": 5.1977,
|
|
"mean_token_accuracy": 0.17872464656829834,
|
|
"num_tokens": 6973933.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.437633657455445,
|
|
"epoch": 3.065424806905952,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.0004402705380661302,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.17672280818223954,
|
|
"num_tokens": 6984314.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.550196599960327,
|
|
"epoch": 3.0699681962744205,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00044003050782986377,
|
|
"loss": 5.312,
|
|
"mean_token_accuracy": 0.16897685527801515,
|
|
"num_tokens": 6995921.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.630539894104004,
|
|
"epoch": 3.0745115856428895,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004397900703972245,
|
|
"loss": 5.2937,
|
|
"mean_token_accuracy": 0.170492485165596,
|
|
"num_tokens": 7006910.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.5243175506591795,
|
|
"epoch": 3.0790549750113585,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004395492263614677,
|
|
"loss": 5.166,
|
|
"mean_token_accuracy": 0.17316992282867433,
|
|
"num_tokens": 7017229.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.532933950424194,
|
|
"epoch": 3.0835983643798275,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043930797631685234,
|
|
"loss": 5.2619,
|
|
"mean_token_accuracy": 0.17371898293495178,
|
|
"num_tokens": 7027934.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.5187013149261475,
|
|
"epoch": 3.088141753748296,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00043906632085863876,
|
|
"loss": 5.1739,
|
|
"mean_token_accuracy": 0.17809343934059144,
|
|
"num_tokens": 7037964.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.540422010421753,
|
|
"epoch": 3.092685143116765,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043882426058308775,
|
|
"loss": 5.1946,
|
|
"mean_token_accuracy": 0.1763277605175972,
|
|
"num_tokens": 7048275.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.471914529800415,
|
|
"epoch": 3.097228532485234,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00043858179608745915,
|
|
"loss": 5.2035,
|
|
"mean_token_accuracy": 0.1735120266675949,
|
|
"num_tokens": 7058488.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.574077606201172,
|
|
"epoch": 3.101771921853703,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004383389279700098,
|
|
"loss": 5.2382,
|
|
"mean_token_accuracy": 0.17403238117694855,
|
|
"num_tokens": 7069195.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.495459032058716,
|
|
"epoch": 3.1063153112221715,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00043809565682999274,
|
|
"loss": 5.1664,
|
|
"mean_token_accuracy": 0.17585339099168779,
|
|
"num_tokens": 7078858.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.472033739089966,
|
|
"epoch": 3.1108587005906405,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00043785198326765535,
|
|
"loss": 5.2061,
|
|
"mean_token_accuracy": 0.17319677025079727,
|
|
"num_tokens": 7088551.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.483696031570434,
|
|
"epoch": 3.1154020899591095,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00043760790788423785,
|
|
"loss": 5.1836,
|
|
"mean_token_accuracy": 0.18097377121448516,
|
|
"num_tokens": 7098987.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.475705194473266,
|
|
"epoch": 3.1199454793275785,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043736343128197207,
|
|
"loss": 5.195,
|
|
"mean_token_accuracy": 0.17989107817411423,
|
|
"num_tokens": 7108337.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.507874584197998,
|
|
"epoch": 3.124488868696047,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004371185540640796,
|
|
"loss": 5.2332,
|
|
"mean_token_accuracy": 0.1778836131095886,
|
|
"num_tokens": 7118267.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.508310413360595,
|
|
"epoch": 3.129032258064516,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00043687327683477077,
|
|
"loss": 5.2724,
|
|
"mean_token_accuracy": 0.17600389271974565,
|
|
"num_tokens": 7129780.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.481181001663208,
|
|
"epoch": 3.133575647432985,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004366276001992427,
|
|
"loss": 5.2253,
|
|
"mean_token_accuracy": 0.1728798419237137,
|
|
"num_tokens": 7139581.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.502482795715332,
|
|
"epoch": 3.138119036801454,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000436381524763678,
|
|
"loss": 5.2064,
|
|
"mean_token_accuracy": 0.17794367671012878,
|
|
"num_tokens": 7150093.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.512056636810303,
|
|
"epoch": 3.142662426169923,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004361350511352435,
|
|
"loss": 5.1546,
|
|
"mean_token_accuracy": 0.17750439494848252,
|
|
"num_tokens": 7159668.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.5208704471588135,
|
|
"epoch": 3.1472058155383915,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004358881799220882,
|
|
"loss": 5.2479,
|
|
"mean_token_accuracy": 0.17130112051963806,
|
|
"num_tokens": 7169713.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.461503219604492,
|
|
"epoch": 3.1517492049068605,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00043564091173334236,
|
|
"loss": 5.2274,
|
|
"mean_token_accuracy": 0.17079599052667618,
|
|
"num_tokens": 7180681.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.476870107650757,
|
|
"epoch": 3.1562925942753295,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00043539324717911573,
|
|
"loss": 5.1048,
|
|
"mean_token_accuracy": 0.18265012800693511,
|
|
"num_tokens": 7189922.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.488419866561889,
|
|
"epoch": 3.1608359836437985,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043514518687049587,
|
|
"loss": 5.1901,
|
|
"mean_token_accuracy": 0.18352550715208055,
|
|
"num_tokens": 7200388.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.485075950622559,
|
|
"epoch": 3.165379373012267,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000434896731419547,
|
|
"loss": 5.2307,
|
|
"mean_token_accuracy": 0.17967755049467088,
|
|
"num_tokens": 7211219.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.404648113250732,
|
|
"epoch": 3.169922762380736,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004346478814393082,
|
|
"loss": 5.0352,
|
|
"mean_token_accuracy": 0.187244413793087,
|
|
"num_tokens": 7220585.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.471985101699829,
|
|
"epoch": 3.174466151749205,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004343986375437921,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.1802661970257759,
|
|
"num_tokens": 7231123.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.510929727554322,
|
|
"epoch": 3.179009541117674,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043414900034798314,
|
|
"loss": 5.1987,
|
|
"mean_token_accuracy": 0.17463556230068206,
|
|
"num_tokens": 7241802.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.179009541117674,
|
|
"eval_entropy": 5.382433459925097,
|
|
"eval_loss": 5.899746894836426,
|
|
"eval_mean_token_accuracy": 0.15575186327792878,
|
|
"eval_num_tokens": 7241802.0,
|
|
"eval_runtime": 2.595,
|
|
"eval_samples_per_second": 1323.696,
|
|
"eval_steps_per_second": 165.703,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.509333992004395,
|
|
"epoch": 3.1835529304861425,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004338989704678364,
|
|
"loss": 5.2603,
|
|
"mean_token_accuracy": 0.17249584794044495,
|
|
"num_tokens": 7252014.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.495342588424682,
|
|
"epoch": 3.1880963198546115,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00043364854852027566,
|
|
"loss": 5.1921,
|
|
"mean_token_accuracy": 0.18153546154499053,
|
|
"num_tokens": 7261738.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.5119648456573485,
|
|
"epoch": 3.1926397092230805,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004333977351231924,
|
|
"loss": 5.2549,
|
|
"mean_token_accuracy": 0.1764304980635643,
|
|
"num_tokens": 7272342.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.4228600025177,
|
|
"epoch": 3.1971830985915495,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00043314653089544336,
|
|
"loss": 5.1519,
|
|
"mean_token_accuracy": 0.18419952392578126,
|
|
"num_tokens": 7282519.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.469780540466308,
|
|
"epoch": 3.201726487960018,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004328949364568504,
|
|
"loss": 5.2388,
|
|
"mean_token_accuracy": 0.18364376723766326,
|
|
"num_tokens": 7292675.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.574817323684693,
|
|
"epoch": 3.206269877328487,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004326429524281975,
|
|
"loss": 5.2772,
|
|
"mean_token_accuracy": 0.17021143883466722,
|
|
"num_tokens": 7302220.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.5049505710601805,
|
|
"epoch": 3.210813266696956,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004323905794312303,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.17765224277973174,
|
|
"num_tokens": 7311942.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.453378772735595,
|
|
"epoch": 3.215356656065425,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00043213781808865414,
|
|
"loss": 5.1307,
|
|
"mean_token_accuracy": 0.18642832487821578,
|
|
"num_tokens": 7322083.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.398830842971802,
|
|
"epoch": 3.2199000454338935,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004318846690241324,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.1770736411213875,
|
|
"num_tokens": 7332291.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.444634485244751,
|
|
"epoch": 3.2244434348023625,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00043163113286228533,
|
|
"loss": 5.0989,
|
|
"mean_token_accuracy": 0.18064910918474197,
|
|
"num_tokens": 7341702.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.446562576293945,
|
|
"epoch": 3.2289868241708315,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004313772102286882,
|
|
"loss": 5.1671,
|
|
"mean_token_accuracy": 0.17366615682840347,
|
|
"num_tokens": 7353332.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.376713371276855,
|
|
"epoch": 3.2335302135393005,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004311229017498698,
|
|
"loss": 5.1392,
|
|
"mean_token_accuracy": 0.18408064693212509,
|
|
"num_tokens": 7362965.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.494131994247437,
|
|
"epoch": 3.238073602907769,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000430868208053311,
|
|
"loss": 5.2489,
|
|
"mean_token_accuracy": 0.16785621792078018,
|
|
"num_tokens": 7374621.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.560452461242676,
|
|
"epoch": 3.242616992276238,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043061312976744326,
|
|
"loss": 5.2703,
|
|
"mean_token_accuracy": 0.17512461245059968,
|
|
"num_tokens": 7385826.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.482319974899292,
|
|
"epoch": 3.247160381644707,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004303576675216469,
|
|
"loss": 5.2997,
|
|
"mean_token_accuracy": 0.16780597418546678,
|
|
"num_tokens": 7397136.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.436460399627686,
|
|
"epoch": 3.251703771013176,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043010182194624957,
|
|
"loss": 5.1574,
|
|
"mean_token_accuracy": 0.17483337968587875,
|
|
"num_tokens": 7407126.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.481370067596435,
|
|
"epoch": 3.2562471603816445,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004298455936725248,
|
|
"loss": 5.169,
|
|
"mean_token_accuracy": 0.17819081842899323,
|
|
"num_tokens": 7416719.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.434046840667724,
|
|
"epoch": 3.2607905497501135,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004295889833326903,
|
|
"loss": 5.2208,
|
|
"mean_token_accuracy": 0.18038192838430406,
|
|
"num_tokens": 7426428.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.431798887252808,
|
|
"epoch": 3.2653339391185825,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00042933199155990675,
|
|
"loss": 5.2008,
|
|
"mean_token_accuracy": 0.1804846629500389,
|
|
"num_tokens": 7436877.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.591900300979614,
|
|
"epoch": 3.2698773284870515,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004290746189882757,
|
|
"loss": 5.2678,
|
|
"mean_token_accuracy": 0.17817400097846986,
|
|
"num_tokens": 7447726.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.479192543029785,
|
|
"epoch": 3.27442071785552,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042881686625283844,
|
|
"loss": 5.2354,
|
|
"mean_token_accuracy": 0.17252737581729888,
|
|
"num_tokens": 7458201.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.539105939865112,
|
|
"epoch": 3.278964107223989,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004285587339895742,
|
|
"loss": 5.2454,
|
|
"mean_token_accuracy": 0.1768377184867859,
|
|
"num_tokens": 7467969.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.501927900314331,
|
|
"epoch": 3.283507496592458,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042830022283539863,
|
|
"loss": 5.2868,
|
|
"mean_token_accuracy": 0.17446671426296234,
|
|
"num_tokens": 7478545.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.547687959671021,
|
|
"epoch": 3.288050885960927,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004280413334281623,
|
|
"loss": 5.2116,
|
|
"mean_token_accuracy": 0.18183978348970414,
|
|
"num_tokens": 7488318.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.582969236373901,
|
|
"epoch": 3.292594275329396,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00042778206640664915,
|
|
"loss": 5.238,
|
|
"mean_token_accuracy": 0.1818904995918274,
|
|
"num_tokens": 7497964.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.471549844741821,
|
|
"epoch": 3.2971376646978645,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004275224224105748,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.18141870945692062,
|
|
"num_tokens": 7508288.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.381367349624634,
|
|
"epoch": 3.3016810540663335,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00042726240208058484,
|
|
"loss": 5.2459,
|
|
"mean_token_accuracy": 0.1785225048661232,
|
|
"num_tokens": 7518716.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.5690727710723875,
|
|
"epoch": 3.3062244434348025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004270020060582538,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.17897082567214967,
|
|
"num_tokens": 7529165.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.421682453155517,
|
|
"epoch": 3.310767832803271,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042674123498608285,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.17114780843257904,
|
|
"num_tokens": 7539975.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.446464872360229,
|
|
"epoch": 3.31531122217174,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00042648008950749877,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.1804968684911728,
|
|
"num_tokens": 7549997.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.46071457862854,
|
|
"epoch": 3.319854611540209,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000426218570266852,
|
|
"loss": 5.2228,
|
|
"mean_token_accuracy": 0.17592016458511353,
|
|
"num_tokens": 7559631.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.386192321777344,
|
|
"epoch": 3.324398000908678,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00042595667790941536,
|
|
"loss": 5.1645,
|
|
"mean_token_accuracy": 0.177582186460495,
|
|
"num_tokens": 7569523.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.422586727142334,
|
|
"epoch": 3.328941390277147,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00042569441308138204,
|
|
"loss": 5.1504,
|
|
"mean_token_accuracy": 0.18387896120548247,
|
|
"num_tokens": 7580339.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.457742643356323,
|
|
"epoch": 3.3334847796456155,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004254317764298645,
|
|
"loss": 5.2071,
|
|
"mean_token_accuracy": 0.1804758444428444,
|
|
"num_tokens": 7591265.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.446174669265747,
|
|
"epoch": 3.3380281690140845,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004251687686028927,
|
|
"loss": 5.2518,
|
|
"mean_token_accuracy": 0.17855857610702514,
|
|
"num_tokens": 7600518.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.550924682617188,
|
|
"epoch": 3.3425715583825535,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042490539024941216,
|
|
"loss": 5.3545,
|
|
"mean_token_accuracy": 0.17275350093841552,
|
|
"num_tokens": 7611568.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.468852901458741,
|
|
"epoch": 3.347114947751022,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00042464164201928286,
|
|
"loss": 5.2871,
|
|
"mean_token_accuracy": 0.17705188244581221,
|
|
"num_tokens": 7621514.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.496052980422974,
|
|
"epoch": 3.351658337119491,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00042437752456327734,
|
|
"loss": 5.2332,
|
|
"mean_token_accuracy": 0.17870181202888488,
|
|
"num_tokens": 7632249.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.469215393066406,
|
|
"epoch": 3.35620172648796,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00042411303853307907,
|
|
"loss": 5.2463,
|
|
"mean_token_accuracy": 0.17592437714338302,
|
|
"num_tokens": 7643539.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.467325019836426,
|
|
"epoch": 3.360745115856429,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004238481845812811,
|
|
"loss": 5.2359,
|
|
"mean_token_accuracy": 0.17237138897180557,
|
|
"num_tokens": 7654089.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.5190749168396,
|
|
"epoch": 3.365288505224898,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042358296336138436,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.1816728428006172,
|
|
"num_tokens": 7665421.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.44252576828003,
|
|
"epoch": 3.3698318945933665,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004233173755277956,
|
|
"loss": 5.2276,
|
|
"mean_token_accuracy": 0.17580224722623825,
|
|
"num_tokens": 7675971.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.437962865829467,
|
|
"epoch": 3.3743752839618355,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042305142173582657,
|
|
"loss": 5.1978,
|
|
"mean_token_accuracy": 0.18061259835958482,
|
|
"num_tokens": 7687444.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.51935977935791,
|
|
"epoch": 3.3789186733303045,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00042278510264169165,
|
|
"loss": 5.262,
|
|
"mean_token_accuracy": 0.17573194205760956,
|
|
"num_tokens": 7697700.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.447100496292114,
|
|
"epoch": 3.3834620626987735,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042251841890250696,
|
|
"loss": 5.1537,
|
|
"mean_token_accuracy": 0.1750078782439232,
|
|
"num_tokens": 7708118.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.451050853729248,
|
|
"epoch": 3.388005452067242,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000422251371176288,
|
|
"loss": 5.1888,
|
|
"mean_token_accuracy": 0.17773102074861527,
|
|
"num_tokens": 7718558.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.523703527450562,
|
|
"epoch": 3.392548841435711,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00042198396012194854,
|
|
"loss": 5.2088,
|
|
"mean_token_accuracy": 0.18126829117536544,
|
|
"num_tokens": 7727873.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.539982032775879,
|
|
"epoch": 3.39709223080418,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00042171618639929865,
|
|
"loss": 5.2754,
|
|
"mean_token_accuracy": 0.17388530671596528,
|
|
"num_tokens": 7737885.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.466766881942749,
|
|
"epoch": 3.401635620172649,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042144805066904345,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.1796223446726799,
|
|
"num_tokens": 7748423.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.495293712615966,
|
|
"epoch": 3.4061790095411175,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042117955359278125,
|
|
"loss": 5.2911,
|
|
"mean_token_accuracy": 0.18074566423892974,
|
|
"num_tokens": 7758879.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.513649415969849,
|
|
"epoch": 3.4107223989095865,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042091069583300177,
|
|
"loss": 5.276,
|
|
"mean_token_accuracy": 0.1744708687067032,
|
|
"num_tokens": 7768765.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.434411239624024,
|
|
"epoch": 3.4152657882780555,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004206414780530849,
|
|
"loss": 5.1779,
|
|
"mean_token_accuracy": 0.1791539743542671,
|
|
"num_tokens": 7778680.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.466006469726563,
|
|
"epoch": 3.4198091776465245,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004203719009172987,
|
|
"loss": 5.2128,
|
|
"mean_token_accuracy": 0.1752891406416893,
|
|
"num_tokens": 7787423.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.498023223876953,
|
|
"epoch": 3.424352567014993,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000420101965090798,
|
|
"loss": 5.2552,
|
|
"mean_token_accuracy": 0.17062118798494338,
|
|
"num_tokens": 7797358.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.546057319641113,
|
|
"epoch": 3.428895956383462,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004198316712396226,
|
|
"loss": 5.2402,
|
|
"mean_token_accuracy": 0.17270464897155763,
|
|
"num_tokens": 7807865.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.431520318984985,
|
|
"epoch": 3.433439345751931,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00041956102003069573,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.17134505957365037,
|
|
"num_tokens": 7818363.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.3976071834564205,
|
|
"epoch": 3.4379827351204,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004192900121318223,
|
|
"loss": 5.2204,
|
|
"mean_token_accuracy": 0.18130185455083847,
|
|
"num_tokens": 7827956.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.507646322250366,
|
|
"epoch": 3.4425261244888685,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004190186482116874,
|
|
"loss": 5.3166,
|
|
"mean_token_accuracy": 0.16859505772590638,
|
|
"num_tokens": 7838963.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.4661156177520756,
|
|
"epoch": 3.4470695138573375,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004187469289398545,
|
|
"loss": 5.1773,
|
|
"mean_token_accuracy": 0.1807806134223938,
|
|
"num_tokens": 7850230.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.389551258087158,
|
|
"epoch": 3.4516129032258065,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041847485498676385,
|
|
"loss": 5.1249,
|
|
"mean_token_accuracy": 0.18048952966928483,
|
|
"num_tokens": 7861514.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.443158769607544,
|
|
"epoch": 3.4561562925942755,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00041820242702373094,
|
|
"loss": 5.2346,
|
|
"mean_token_accuracy": 0.18045746237039567,
|
|
"num_tokens": 7871175.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.463071680068969,
|
|
"epoch": 3.460699681962744,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004179296457229448,
|
|
"loss": 5.2574,
|
|
"mean_token_accuracy": 0.17358458638191224,
|
|
"num_tokens": 7882312.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.44131441116333,
|
|
"epoch": 3.465243071331213,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004176565117574659,
|
|
"loss": 5.1794,
|
|
"mean_token_accuracy": 0.1781587168574333,
|
|
"num_tokens": 7892904.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.4369011402130125,
|
|
"epoch": 3.469786460699682,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004173830258012255,
|
|
"loss": 5.246,
|
|
"mean_token_accuracy": 0.17424970269203185,
|
|
"num_tokens": 7903127.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.490753078460694,
|
|
"epoch": 3.474329850068151,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004171091885290226,
|
|
"loss": 5.2609,
|
|
"mean_token_accuracy": 0.17753191739320756,
|
|
"num_tokens": 7914041.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.4483214855194095,
|
|
"epoch": 3.4788732394366195,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004168350006165238,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.17027440518140793,
|
|
"num_tokens": 7924334.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.53763976097107,
|
|
"epoch": 3.4834166288050885,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00041656046274026034,
|
|
"loss": 5.2744,
|
|
"mean_token_accuracy": 0.17467211037874222,
|
|
"num_tokens": 7935973.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.549361228942871,
|
|
"epoch": 3.4879600181735575,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004162855755776272,
|
|
"loss": 5.2338,
|
|
"mean_token_accuracy": 0.17483218908309936,
|
|
"num_tokens": 7946780.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.463102865219116,
|
|
"epoch": 3.4925034075420265,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00041601033980688106,
|
|
"loss": 5.294,
|
|
"mean_token_accuracy": 0.17409658282995225,
|
|
"num_tokens": 7957243.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.648853969573975,
|
|
"epoch": 3.497046796910495,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004157347561071389,
|
|
"loss": 5.3908,
|
|
"mean_token_accuracy": 0.17060235142707825,
|
|
"num_tokens": 7968160.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.447634172439575,
|
|
"epoch": 3.501590186278964,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00041545882515837604,
|
|
"loss": 5.264,
|
|
"mean_token_accuracy": 0.16583557426929474,
|
|
"num_tokens": 7977910.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.4974767684936525,
|
|
"epoch": 3.506133575647433,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00041518254764142477,
|
|
"loss": 5.2357,
|
|
"mean_token_accuracy": 0.1763804405927658,
|
|
"num_tokens": 7988239.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.492964410781861,
|
|
"epoch": 3.510676965015902,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041490592423797226,
|
|
"loss": 5.1922,
|
|
"mean_token_accuracy": 0.1785515993833542,
|
|
"num_tokens": 7998128.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.3743854522705075,
|
|
"epoch": 3.515220354384371,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004146289556305594,
|
|
"loss": 5.191,
|
|
"mean_token_accuracy": 0.1789640963077545,
|
|
"num_tokens": 8008997.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.446508359909058,
|
|
"epoch": 3.5197637437528395,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004143516425025786,
|
|
"loss": 5.2634,
|
|
"mean_token_accuracy": 0.17850781679153443,
|
|
"num_tokens": 8018864.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.536286115646362,
|
|
"epoch": 3.5243071331213085,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004140739855382724,
|
|
"loss": 5.2819,
|
|
"mean_token_accuracy": 0.17064804583787918,
|
|
"num_tokens": 8029420.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.510391330718994,
|
|
"epoch": 3.5288505224897775,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00041379598542273197,
|
|
"loss": 5.3342,
|
|
"mean_token_accuracy": 0.17117436081171036,
|
|
"num_tokens": 8040184.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.542795991897583,
|
|
"epoch": 3.533393911858246,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00041351764284189464,
|
|
"loss": 5.2409,
|
|
"mean_token_accuracy": 0.17732380330562592,
|
|
"num_tokens": 8050291.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.454309320449829,
|
|
"epoch": 3.537937301226715,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004132389584825433,
|
|
"loss": 5.2288,
|
|
"mean_token_accuracy": 0.174114890396595,
|
|
"num_tokens": 8060399.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.411073017120361,
|
|
"epoch": 3.542480690595184,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00041295993303230383,
|
|
"loss": 5.2265,
|
|
"mean_token_accuracy": 0.18468950986862182,
|
|
"num_tokens": 8070986.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.515261507034301,
|
|
"epoch": 3.547024079963653,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004126805671796438,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.18100871443748473,
|
|
"num_tokens": 8080872.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.545407247543335,
|
|
"epoch": 3.551567469332122,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004124008616138707,
|
|
"loss": 5.2936,
|
|
"mean_token_accuracy": 0.175253963470459,
|
|
"num_tokens": 8091618.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.476496839523316,
|
|
"epoch": 3.5561108587005905,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00041212081702513015,
|
|
"loss": 5.2342,
|
|
"mean_token_accuracy": 0.17120363414287568,
|
|
"num_tokens": 8101904.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.434841728210449,
|
|
"epoch": 3.5606542480690595,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004118404341044044,
|
|
"loss": 5.1718,
|
|
"mean_token_accuracy": 0.18606000691652297,
|
|
"num_tokens": 8112879.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.422662353515625,
|
|
"epoch": 3.5651976374375285,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00041155971354351034,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.18221483677625655,
|
|
"num_tokens": 8123104.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.41081428527832,
|
|
"epoch": 3.569741026805997,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00041127865603509815,
|
|
"loss": 5.2128,
|
|
"mean_token_accuracy": 0.1819761112332344,
|
|
"num_tokens": 8132983.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.446842098236084,
|
|
"epoch": 3.574284416174466,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004109972622726493,
|
|
"loss": 5.1731,
|
|
"mean_token_accuracy": 0.17434430718421937,
|
|
"num_tokens": 8143344.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.503196096420288,
|
|
"epoch": 3.578827805542935,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041071553295047485,
|
|
"loss": 5.2275,
|
|
"mean_token_accuracy": 0.1748381093144417,
|
|
"num_tokens": 8153561.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.457211256027222,
|
|
"epoch": 3.583371194911404,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004104334687637139,
|
|
"loss": 5.1814,
|
|
"mean_token_accuracy": 0.18217648416757584,
|
|
"num_tokens": 8163713.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.37618579864502,
|
|
"epoch": 3.587914584279873,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00041015107040833187,
|
|
"loss": 5.1339,
|
|
"mean_token_accuracy": 0.18402476310729982,
|
|
"num_tokens": 8174540.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.393379974365234,
|
|
"epoch": 3.5924579736483415,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004098683385811185,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.17773871421813964,
|
|
"num_tokens": 8185254.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.55565333366394,
|
|
"epoch": 3.5970013630168105,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00040958527397968666,
|
|
"loss": 5.2425,
|
|
"mean_token_accuracy": 0.17734598666429519,
|
|
"num_tokens": 8195205.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.485688066482544,
|
|
"epoch": 3.6015447523852795,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004093018773024699,
|
|
"loss": 5.2858,
|
|
"mean_token_accuracy": 0.1731982946395874,
|
|
"num_tokens": 8205298.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.3778839111328125,
|
|
"epoch": 3.606088141753748,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 0.0004090181492487215,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.18281037658452987,
|
|
"num_tokens": 8215677.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.374654626846313,
|
|
"epoch": 3.610631531122217,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00040873409051851223,
|
|
"loss": 5.1622,
|
|
"mean_token_accuracy": 0.17445134222507477,
|
|
"num_tokens": 8226938.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.485167455673218,
|
|
"epoch": 3.615174920490686,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004084497018127286,
|
|
"loss": 5.23,
|
|
"mean_token_accuracy": 0.1727091148495674,
|
|
"num_tokens": 8238718.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.426266288757324,
|
|
"epoch": 3.619718309859155,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004081649838330717,
|
|
"loss": 5.1841,
|
|
"mean_token_accuracy": 0.17510856688022614,
|
|
"num_tokens": 8249562.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.479261064529419,
|
|
"epoch": 3.624261699227624,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00040787993728205464,
|
|
"loss": 5.2753,
|
|
"mean_token_accuracy": 0.174649153649807,
|
|
"num_tokens": 8259736.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.522538328170777,
|
|
"epoch": 3.6288050885960925,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00040759456286300154,
|
|
"loss": 5.225,
|
|
"mean_token_accuracy": 0.17999522536993026,
|
|
"num_tokens": 8269933.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.495010757446289,
|
|
"epoch": 3.6333484779645615,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040730886128004544,
|
|
"loss": 5.258,
|
|
"mean_token_accuracy": 0.17528275847435,
|
|
"num_tokens": 8279950.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.6333484779645615,
|
|
"eval_entropy": 5.41199811336606,
|
|
"eval_loss": 5.862250328063965,
|
|
"eval_mean_token_accuracy": 0.15924781813871028,
|
|
"eval_num_tokens": 8279950.0,
|
|
"eval_runtime": 2.5918,
|
|
"eval_samples_per_second": 1325.356,
|
|
"eval_steps_per_second": 165.911,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.521580028533935,
|
|
"epoch": 3.6378918673330305,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004070228332381266,
|
|
"loss": 5.2308,
|
|
"mean_token_accuracy": 0.1797862634062767,
|
|
"num_tokens": 8289575.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.4236414432525635,
|
|
"epoch": 3.6424352567014995,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004067364794429908,
|
|
"loss": 5.2479,
|
|
"mean_token_accuracy": 0.17372748553752898,
|
|
"num_tokens": 8300809.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.497612762451172,
|
|
"epoch": 3.646978646069968,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004064498006011874,
|
|
"loss": 5.242,
|
|
"mean_token_accuracy": 0.17624309211969375,
|
|
"num_tokens": 8310666.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.440974330902099,
|
|
"epoch": 3.651522035438437,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00040616279742006813,
|
|
"loss": 5.2074,
|
|
"mean_token_accuracy": 0.17368254214525222,
|
|
"num_tokens": 8320823.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.476006603240966,
|
|
"epoch": 3.656065424806906,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00040587547060778466,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.175482314825058,
|
|
"num_tokens": 8330343.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.388587379455567,
|
|
"epoch": 3.660608814175375,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004055878208732874,
|
|
"loss": 5.1531,
|
|
"mean_token_accuracy": 0.1859850212931633,
|
|
"num_tokens": 8341444.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.448707580566406,
|
|
"epoch": 3.665152203543844,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00040529984892632363,
|
|
"loss": 5.2565,
|
|
"mean_token_accuracy": 0.172608782351017,
|
|
"num_tokens": 8351654.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.496041917800904,
|
|
"epoch": 3.6696955929123125,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004050115554774353,
|
|
"loss": 5.24,
|
|
"mean_token_accuracy": 0.17165805548429489,
|
|
"num_tokens": 8361534.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.450342512130737,
|
|
"epoch": 3.6742389822807815,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004047229412379578,
|
|
"loss": 5.2368,
|
|
"mean_token_accuracy": 0.18036909401416779,
|
|
"num_tokens": 8371955.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.48595929145813,
|
|
"epoch": 3.6787823716492505,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004044340069200183,
|
|
"loss": 5.2581,
|
|
"mean_token_accuracy": 0.17732932418584824,
|
|
"num_tokens": 8383014.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.463910484313965,
|
|
"epoch": 3.683325761017719,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00040414475323653326,
|
|
"loss": 5.2285,
|
|
"mean_token_accuracy": 0.1790663257241249,
|
|
"num_tokens": 8393997.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.448544120788574,
|
|
"epoch": 3.687869150386188,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004038551809012076,
|
|
"loss": 5.2018,
|
|
"mean_token_accuracy": 0.17442853301763533,
|
|
"num_tokens": 8404712.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.522753286361694,
|
|
"epoch": 3.692412539754657,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004035652906285322,
|
|
"loss": 5.3313,
|
|
"mean_token_accuracy": 0.17518966048955917,
|
|
"num_tokens": 8416610.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.436816692352295,
|
|
"epoch": 3.696955929123126,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00040327508313378244,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.18068812042474747,
|
|
"num_tokens": 8427096.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.523792028427124,
|
|
"epoch": 3.701499318491595,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004029845591330164,
|
|
"loss": 5.2033,
|
|
"mean_token_accuracy": 0.17596746385097503,
|
|
"num_tokens": 8438748.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.533337783813477,
|
|
"epoch": 3.7060427078600635,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040269371934307336,
|
|
"loss": 5.2954,
|
|
"mean_token_accuracy": 0.1742219641804695,
|
|
"num_tokens": 8449293.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.491851186752319,
|
|
"epoch": 3.7105860972285325,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040240256448157146,
|
|
"loss": 5.2094,
|
|
"mean_token_accuracy": 0.17826563119888306,
|
|
"num_tokens": 8459465.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.3955024719238285,
|
|
"epoch": 3.7151294865970015,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004021110952669065,
|
|
"loss": 5.2349,
|
|
"mean_token_accuracy": 0.17518757432699203,
|
|
"num_tokens": 8469589.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.425784015655518,
|
|
"epoch": 3.71967287596547,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004018193124182497,
|
|
"loss": 5.2084,
|
|
"mean_token_accuracy": 0.1720123589038849,
|
|
"num_tokens": 8480201.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.429391717910766,
|
|
"epoch": 3.724216265333939,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004015272166555462,
|
|
"loss": 5.2239,
|
|
"mean_token_accuracy": 0.1794332191348076,
|
|
"num_tokens": 8492495.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.466448450088501,
|
|
"epoch": 3.728759654702408,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004012348086995132,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.1759302794933319,
|
|
"num_tokens": 8503429.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.552805089950562,
|
|
"epoch": 3.733303044070877,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004009420892716385,
|
|
"loss": 5.2519,
|
|
"mean_token_accuracy": 0.17492673099040984,
|
|
"num_tokens": 8513169.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.440295886993408,
|
|
"epoch": 3.737846433439346,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004006490590941781,
|
|
"loss": 5.1842,
|
|
"mean_token_accuracy": 0.17969796806573868,
|
|
"num_tokens": 8523192.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.38798303604126,
|
|
"epoch": 3.7423898228078145,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004003557188901547,
|
|
"loss": 5.1165,
|
|
"mean_token_accuracy": 0.18241094350814818,
|
|
"num_tokens": 8534410.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.385713577270508,
|
|
"epoch": 3.7469332121762835,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004000620693833562,
|
|
"loss": 5.2554,
|
|
"mean_token_accuracy": 0.18005164414644242,
|
|
"num_tokens": 8544317.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.416796064376831,
|
|
"epoch": 3.7514766015447525,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003997681112983335,
|
|
"loss": 5.1284,
|
|
"mean_token_accuracy": 0.17902693748474122,
|
|
"num_tokens": 8554730.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.416399765014648,
|
|
"epoch": 3.756019990913221,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00039947384536039905,
|
|
"loss": 5.1933,
|
|
"mean_token_accuracy": 0.17964504659175873,
|
|
"num_tokens": 8564909.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.485965347290039,
|
|
"epoch": 3.76056338028169,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039917927229562474,
|
|
"loss": 5.1983,
|
|
"mean_token_accuracy": 0.18036925047636032,
|
|
"num_tokens": 8575290.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.479689979553223,
|
|
"epoch": 3.765106769650159,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039888439283084035,
|
|
"loss": 5.2426,
|
|
"mean_token_accuracy": 0.18220258504152298,
|
|
"num_tokens": 8585521.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.414248371124268,
|
|
"epoch": 3.769650159018628,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039858920769363164,
|
|
"loss": 5.1381,
|
|
"mean_token_accuracy": 0.17874915897846222,
|
|
"num_tokens": 8595049.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.408684301376343,
|
|
"epoch": 3.774193548387097,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003982937176123386,
|
|
"loss": 5.2333,
|
|
"mean_token_accuracy": 0.17589736729860306,
|
|
"num_tokens": 8604707.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.5225732803344725,
|
|
"epoch": 3.7787369377555655,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039799792331605364,
|
|
"loss": 5.193,
|
|
"mean_token_accuracy": 0.17463405877351762,
|
|
"num_tokens": 8615573.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.454258680343628,
|
|
"epoch": 3.7832803271240345,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00039770182553461965,
|
|
"loss": 5.2245,
|
|
"mean_token_accuracy": 0.17947518080472946,
|
|
"num_tokens": 8625799.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.435826969146729,
|
|
"epoch": 3.7878237164925035,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00039740542499862866,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.17867654412984849,
|
|
"num_tokens": 8635351.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.471651363372803,
|
|
"epoch": 3.792367105860972,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003971087224394194,
|
|
"loss": 5.3034,
|
|
"mean_token_accuracy": 0.175942163169384,
|
|
"num_tokens": 8645765.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.516590595245361,
|
|
"epoch": 3.796910495229441,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003968117185890761,
|
|
"loss": 5.249,
|
|
"mean_token_accuracy": 0.17524760514497756,
|
|
"num_tokens": 8656170.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.433881044387817,
|
|
"epoch": 3.80145388459791,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039651441418042606,
|
|
"loss": 5.1604,
|
|
"mean_token_accuracy": 0.18399293720722198,
|
|
"num_tokens": 8665704.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.456085681915283,
|
|
"epoch": 3.805997273966379,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003962168099470384,
|
|
"loss": 5.2421,
|
|
"mean_token_accuracy": 0.17985902428627015,
|
|
"num_tokens": 8676236.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.52960729598999,
|
|
"epoch": 3.810540663334848,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000395918906623222,
|
|
"loss": 5.2644,
|
|
"mean_token_accuracy": 0.17919850945472718,
|
|
"num_tokens": 8687242.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.448080253601074,
|
|
"epoch": 3.8150840527033165,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.0003956207049440238,
|
|
"loss": 5.2384,
|
|
"mean_token_accuracy": 0.18030489534139632,
|
|
"num_tokens": 8696967.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.42124924659729,
|
|
"epoch": 3.8196274420717855,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003953222056452265,
|
|
"loss": 5.1594,
|
|
"mean_token_accuracy": 0.18724761307239532,
|
|
"num_tokens": 8707832.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.366513061523437,
|
|
"epoch": 3.8241708314402545,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003950234094633479,
|
|
"loss": 5.2179,
|
|
"mean_token_accuracy": 0.17358993440866471,
|
|
"num_tokens": 8717592.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.484923315048218,
|
|
"epoch": 3.828714220808723,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00039472431713563746,
|
|
"loss": 5.251,
|
|
"mean_token_accuracy": 0.17891739010810853,
|
|
"num_tokens": 8728899.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.444270515441895,
|
|
"epoch": 3.833257610177192,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000394424929400076,
|
|
"loss": 5.249,
|
|
"mean_token_accuracy": 0.17359264492988585,
|
|
"num_tokens": 8740092.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.478668880462647,
|
|
"epoch": 3.837800999545661,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003941252469953729,
|
|
"loss": 5.227,
|
|
"mean_token_accuracy": 0.17877087444067002,
|
|
"num_tokens": 8750256.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.419363927841187,
|
|
"epoch": 3.84234438891413,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003938252706609647,
|
|
"loss": 5.2154,
|
|
"mean_token_accuracy": 0.1742422178387642,
|
|
"num_tokens": 8760546.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.496589088439942,
|
|
"epoch": 3.846887778282599,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003935250011370132,
|
|
"loss": 5.3173,
|
|
"mean_token_accuracy": 0.17036086022853852,
|
|
"num_tokens": 8770017.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.485684776306153,
|
|
"epoch": 3.851431167651068,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003932244391644037,
|
|
"loss": 5.2282,
|
|
"mean_token_accuracy": 0.17748588621616362,
|
|
"num_tokens": 8780597.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.494361877441406,
|
|
"epoch": 3.8559745570195365,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003929235854847428,
|
|
"loss": 5.2129,
|
|
"mean_token_accuracy": 0.1680305302143097,
|
|
"num_tokens": 8791508.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.487069225311279,
|
|
"epoch": 3.8605179463880055,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039262244084035724,
|
|
"loss": 5.2124,
|
|
"mean_token_accuracy": 0.17978151142597198,
|
|
"num_tokens": 8803013.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.465966606140137,
|
|
"epoch": 3.8650613357564745,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039232100597429143,
|
|
"loss": 5.2172,
|
|
"mean_token_accuracy": 0.1821499526500702,
|
|
"num_tokens": 8812550.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.378421306610107,
|
|
"epoch": 3.869604725124943,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00039201928163030584,
|
|
"loss": 5.1717,
|
|
"mean_token_accuracy": 0.18206015974283218,
|
|
"num_tokens": 8822700.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.470287656784057,
|
|
"epoch": 3.874148114493412,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039171726855287544,
|
|
"loss": 5.2256,
|
|
"mean_token_accuracy": 0.18008975684642792,
|
|
"num_tokens": 8832859.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.413965177536011,
|
|
"epoch": 3.878691503861881,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039141496748718735,
|
|
"loss": 5.1926,
|
|
"mean_token_accuracy": 0.1842578798532486,
|
|
"num_tokens": 8843216.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.443237400054931,
|
|
"epoch": 3.88323489323035,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003911123791791395,
|
|
"loss": 5.1523,
|
|
"mean_token_accuracy": 0.178711000084877,
|
|
"num_tokens": 8854791.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.398017549514771,
|
|
"epoch": 3.887778282598819,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003908095043753384,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.1795601949095726,
|
|
"num_tokens": 8865191.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.4174120903015135,
|
|
"epoch": 3.8923216719672875,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00039050634382309753,
|
|
"loss": 5.1901,
|
|
"mean_token_accuracy": 0.18126186728477478,
|
|
"num_tokens": 8875067.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.424483394622802,
|
|
"epoch": 3.8968650613357565,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003902028982704355,
|
|
"loss": 5.214,
|
|
"mean_token_accuracy": 0.1850364923477173,
|
|
"num_tokens": 8884391.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.456392478942871,
|
|
"epoch": 3.9014084507042255,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003898991684660738,
|
|
"loss": 5.3111,
|
|
"mean_token_accuracy": 0.1736356183886528,
|
|
"num_tokens": 8895077.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.451139354705811,
|
|
"epoch": 3.905951840072694,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003895951551594358,
|
|
"loss": 5.2337,
|
|
"mean_token_accuracy": 0.1822612151503563,
|
|
"num_tokens": 8906754.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.468061017990112,
|
|
"epoch": 3.910495229441163,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000389290859100644,
|
|
"loss": 5.1818,
|
|
"mean_token_accuracy": 0.1806336000561714,
|
|
"num_tokens": 8917839.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.439494466781616,
|
|
"epoch": 3.915038618809632,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003889862810405187,
|
|
"loss": 5.2376,
|
|
"mean_token_accuracy": 0.18658496886491777,
|
|
"num_tokens": 8928555.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.424708986282349,
|
|
"epoch": 3.919582008178101,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000388681421730576,
|
|
"loss": 5.1956,
|
|
"mean_token_accuracy": 0.17549479752779007,
|
|
"num_tokens": 8938811.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.535794448852539,
|
|
"epoch": 3.92412539754657,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003883762819230259,
|
|
"loss": 5.3215,
|
|
"mean_token_accuracy": 0.17117299288511276,
|
|
"num_tokens": 8949302.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.438610649108886,
|
|
"epoch": 3.9286687869150385,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00038807086237077057,
|
|
"loss": 5.2042,
|
|
"mean_token_accuracy": 0.17728144824504852,
|
|
"num_tokens": 8959242.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.4511713027954105,
|
|
"epoch": 3.9332121762835075,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00038776516382740237,
|
|
"loss": 5.2369,
|
|
"mean_token_accuracy": 0.17862620204687119,
|
|
"num_tokens": 8969696.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.3622273921966555,
|
|
"epoch": 3.9377555656519765,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00038745918704720205,
|
|
"loss": 5.2052,
|
|
"mean_token_accuracy": 0.17837002426385878,
|
|
"num_tokens": 8979458.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.419256687164307,
|
|
"epoch": 3.942298955020445,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00038715293278513693,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.17744985073804856,
|
|
"num_tokens": 8989851.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.492688798904419,
|
|
"epoch": 3.946842344388914,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00038684640179685896,
|
|
"loss": 5.3628,
|
|
"mean_token_accuracy": 0.17779286801815034,
|
|
"num_tokens": 9000605.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.488884210586548,
|
|
"epoch": 3.951385733757383,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003865395948387029,
|
|
"loss": 5.1996,
|
|
"mean_token_accuracy": 0.17677636742591857,
|
|
"num_tokens": 9010578.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.475163221359253,
|
|
"epoch": 3.955929123125852,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00038623251266768437,
|
|
"loss": 5.1723,
|
|
"mean_token_accuracy": 0.17941099405288696,
|
|
"num_tokens": 9020697.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.491322326660156,
|
|
"epoch": 3.960472512494321,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00038592515604149815,
|
|
"loss": 5.2416,
|
|
"mean_token_accuracy": 0.1759133070707321,
|
|
"num_tokens": 9030795.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.353063869476318,
|
|
"epoch": 3.9650159018627895,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003856175257185162,
|
|
"loss": 5.2182,
|
|
"mean_token_accuracy": 0.1748987391591072,
|
|
"num_tokens": 9041683.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.420461702346802,
|
|
"epoch": 3.9695592912312585,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00038530962245778564,
|
|
"loss": 5.1737,
|
|
"mean_token_accuracy": 0.18123656511306763,
|
|
"num_tokens": 9051167.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.403790903091431,
|
|
"epoch": 3.9741026805997275,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0003850014470190272,
|
|
"loss": 5.1956,
|
|
"mean_token_accuracy": 0.18561478108167648,
|
|
"num_tokens": 9060514.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.449420499801636,
|
|
"epoch": 3.978646069968196,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0003846930001626332,
|
|
"loss": 5.2241,
|
|
"mean_token_accuracy": 0.17524918168783188,
|
|
"num_tokens": 9070774.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.463682365417481,
|
|
"epoch": 3.983189459336665,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003843842826496656,
|
|
"loss": 5.1348,
|
|
"mean_token_accuracy": 0.1801970988512039,
|
|
"num_tokens": 9080718.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.337631130218506,
|
|
"epoch": 3.987732848705134,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000384075295241854,
|
|
"loss": 5.1507,
|
|
"mean_token_accuracy": 0.1759501338005066,
|
|
"num_tokens": 9091183.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.465139245986938,
|
|
"epoch": 3.992276238073603,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003837660387015944,
|
|
"loss": 5.2078,
|
|
"mean_token_accuracy": 0.1808859348297119,
|
|
"num_tokens": 9101173.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.458251476287842,
|
|
"epoch": 3.996819627442072,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003834565137919463,
|
|
"loss": 5.236,
|
|
"mean_token_accuracy": 0.17722010016441345,
|
|
"num_tokens": 9110724.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.461641046735975,
|
|
"epoch": 4.000908677873694,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0003831467212766318,
|
|
"loss": 5.1955,
|
|
"mean_token_accuracy": 0.18436774611473083,
|
|
"num_tokens": 9120410.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.47430648803711,
|
|
"epoch": 4.005452067242163,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003828366619200331,
|
|
"loss": 5.0542,
|
|
"mean_token_accuracy": 0.18023105561733246,
|
|
"num_tokens": 9130235.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.384516191482544,
|
|
"epoch": 4.009995456610632,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003825263364871908,
|
|
"loss": 4.8353,
|
|
"mean_token_accuracy": 0.2018807828426361,
|
|
"num_tokens": 9140740.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.438038444519043,
|
|
"epoch": 4.0145388459791,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003822157457438022,
|
|
"loss": 4.9927,
|
|
"mean_token_accuracy": 0.19572025835514067,
|
|
"num_tokens": 9151271.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.378810930252075,
|
|
"epoch": 4.019082235347569,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00038190489045621893,
|
|
"loss": 5.0435,
|
|
"mean_token_accuracy": 0.18818851113319396,
|
|
"num_tokens": 9162520.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.441124391555786,
|
|
"epoch": 4.023625624716038,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003815937713914455,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.18008158057928086,
|
|
"num_tokens": 9173920.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.473393249511719,
|
|
"epoch": 4.028169014084507,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00038128238931713743,
|
|
"loss": 4.993,
|
|
"mean_token_accuracy": 0.18537705391645432,
|
|
"num_tokens": 9184507.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.400975751876831,
|
|
"epoch": 4.032712403452976,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00038097074500159883,
|
|
"loss": 5.0104,
|
|
"mean_token_accuracy": 0.18585438579320906,
|
|
"num_tokens": 9194971.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.341420221328735,
|
|
"epoch": 4.037255792821445,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00038065883921378123,
|
|
"loss": 4.8614,
|
|
"mean_token_accuracy": 0.19670613259077072,
|
|
"num_tokens": 9205348.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.441487503051758,
|
|
"epoch": 4.041799182189914,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003803466727232811,
|
|
"loss": 5.1072,
|
|
"mean_token_accuracy": 0.18487307876348497,
|
|
"num_tokens": 9216301.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.37165846824646,
|
|
"epoch": 4.046342571558383,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003800342463003383,
|
|
"loss": 4.9579,
|
|
"mean_token_accuracy": 0.19256972521543503,
|
|
"num_tokens": 9226778.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.413996696472168,
|
|
"epoch": 4.050885960926851,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00037972156071583386,
|
|
"loss": 5.0395,
|
|
"mean_token_accuracy": 0.1883913353085518,
|
|
"num_tokens": 9238309.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.424037837982178,
|
|
"epoch": 4.05542935029532,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003794086167412884,
|
|
"loss": 4.9803,
|
|
"mean_token_accuracy": 0.18984542787075043,
|
|
"num_tokens": 9247720.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.567341041564942,
|
|
"epoch": 4.059972739663789,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003790954151488601,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.19394380152225493,
|
|
"num_tokens": 9257519.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.40732626914978,
|
|
"epoch": 4.064516129032258,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003787819567113427,
|
|
"loss": 5.037,
|
|
"mean_token_accuracy": 0.18419537991285323,
|
|
"num_tokens": 9268554.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.384883642196655,
|
|
"epoch": 4.069059518400727,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003784682422021638,
|
|
"loss": 4.9823,
|
|
"mean_token_accuracy": 0.18527417927980422,
|
|
"num_tokens": 9279354.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.442471885681153,
|
|
"epoch": 4.073602907769196,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0003781542723953828,
|
|
"loss": 5.0096,
|
|
"mean_token_accuracy": 0.1937069669365883,
|
|
"num_tokens": 9290174.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.399314546585083,
|
|
"epoch": 4.078146297137665,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00037784004806568885,
|
|
"loss": 4.9648,
|
|
"mean_token_accuracy": 0.19110300838947297,
|
|
"num_tokens": 9300151.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.307184505462646,
|
|
"epoch": 4.082689686506134,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00037752556998839934,
|
|
"loss": 5.0108,
|
|
"mean_token_accuracy": 0.1916087746620178,
|
|
"num_tokens": 9309868.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.4814074516296385,
|
|
"epoch": 4.087233075874602,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003772108389394576,
|
|
"loss": 5.0124,
|
|
"mean_token_accuracy": 0.18615510016679765,
|
|
"num_tokens": 9319753.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.087233075874602,
|
|
"eval_entropy": 5.347864997109702,
|
|
"eval_loss": 5.8249406814575195,
|
|
"eval_mean_token_accuracy": 0.16103600999990175,
|
|
"eval_num_tokens": 9319753.0,
|
|
"eval_runtime": 2.5929,
|
|
"eval_samples_per_second": 1324.774,
|
|
"eval_steps_per_second": 165.838,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.406263017654419,
|
|
"epoch": 4.091776465243071,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037689585569543137,
|
|
"loss": 4.9617,
|
|
"mean_token_accuracy": 0.19377471059560775,
|
|
"num_tokens": 9330293.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.336021280288696,
|
|
"epoch": 4.09631985461154,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003765806210335103,
|
|
"loss": 4.9502,
|
|
"mean_token_accuracy": 0.19947022050619126,
|
|
"num_tokens": 9340791.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.323797559738159,
|
|
"epoch": 4.100863243980009,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00037626513573150473,
|
|
"loss": 4.9294,
|
|
"mean_token_accuracy": 0.2008962258696556,
|
|
"num_tokens": 9350685.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.400818395614624,
|
|
"epoch": 4.105406633348478,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00037594940056784315,
|
|
"loss": 4.9925,
|
|
"mean_token_accuracy": 0.19366396814584733,
|
|
"num_tokens": 9360474.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.3404755115509035,
|
|
"epoch": 4.109950022716947,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003756334163215708,
|
|
"loss": 5.0087,
|
|
"mean_token_accuracy": 0.18035605996847154,
|
|
"num_tokens": 9371389.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.40869369506836,
|
|
"epoch": 4.114493412085416,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00037531718377234723,
|
|
"loss": 4.9959,
|
|
"mean_token_accuracy": 0.18557582348585128,
|
|
"num_tokens": 9381359.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.417052268981934,
|
|
"epoch": 4.119036801453885,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003750007037004451,
|
|
"loss": 5.0138,
|
|
"mean_token_accuracy": 0.188804467022419,
|
|
"num_tokens": 9391171.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.4005261898040775,
|
|
"epoch": 4.123580190822353,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003746839768867473,
|
|
"loss": 5.0616,
|
|
"mean_token_accuracy": 0.18464367538690568,
|
|
"num_tokens": 9401912.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.405744314193726,
|
|
"epoch": 4.128123580190822,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003743670041127458,
|
|
"loss": 5.0335,
|
|
"mean_token_accuracy": 0.18163667619228363,
|
|
"num_tokens": 9413353.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.4188910007476805,
|
|
"epoch": 4.132666969559291,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003740497861605395,
|
|
"loss": 4.9821,
|
|
"mean_token_accuracy": 0.19036975502967834,
|
|
"num_tokens": 9424236.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.317966175079346,
|
|
"epoch": 4.13721035892776,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000373732323812832,
|
|
"loss": 4.9628,
|
|
"mean_token_accuracy": 0.19187259674072266,
|
|
"num_tokens": 9434291.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.387293910980224,
|
|
"epoch": 4.141753748296229,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00037341461785293025,
|
|
"loss": 5.0376,
|
|
"mean_token_accuracy": 0.18440647423267365,
|
|
"num_tokens": 9445087.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.3747923374176025,
|
|
"epoch": 4.146297137664698,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003730966690647421,
|
|
"loss": 4.9422,
|
|
"mean_token_accuracy": 0.1965849757194519,
|
|
"num_tokens": 9455479.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.372088193893433,
|
|
"epoch": 4.150840527033167,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00037277847823277444,
|
|
"loss": 5.0406,
|
|
"mean_token_accuracy": 0.18402064293622972,
|
|
"num_tokens": 9467709.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.394093370437622,
|
|
"epoch": 4.155383916401636,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037246004614213164,
|
|
"loss": 5.0494,
|
|
"mean_token_accuracy": 0.17921328991651536,
|
|
"num_tokens": 9478004.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.377377462387085,
|
|
"epoch": 4.159927305770104,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003721413735785132,
|
|
"loss": 4.9089,
|
|
"mean_token_accuracy": 0.19705862700939178,
|
|
"num_tokens": 9486855.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.374736785888672,
|
|
"epoch": 4.164470695138573,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000371822461328212,
|
|
"loss": 5.0371,
|
|
"mean_token_accuracy": 0.1797275260090828,
|
|
"num_tokens": 9497321.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.4494099617004395,
|
|
"epoch": 4.169014084507042,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00037150331017811233,
|
|
"loss": 4.9989,
|
|
"mean_token_accuracy": 0.1932631880044937,
|
|
"num_tokens": 9507299.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.364447069168091,
|
|
"epoch": 4.173557473875511,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003711839209156878,
|
|
"loss": 4.9294,
|
|
"mean_token_accuracy": 0.1987797051668167,
|
|
"num_tokens": 9517800.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.388552236557007,
|
|
"epoch": 4.17810086324398,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003708642943289999,
|
|
"loss": 4.9779,
|
|
"mean_token_accuracy": 0.1958901211619377,
|
|
"num_tokens": 9528215.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.27638578414917,
|
|
"epoch": 4.182644252612449,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003705444312066953,
|
|
"loss": 4.9615,
|
|
"mean_token_accuracy": 0.19837335646152496,
|
|
"num_tokens": 9539515.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.3221760272979735,
|
|
"epoch": 4.187187641980918,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003702243323380045,
|
|
"loss": 5.0151,
|
|
"mean_token_accuracy": 0.189212803542614,
|
|
"num_tokens": 9548709.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.332740831375122,
|
|
"epoch": 4.191731031349387,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00036990399851273965,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.19085367619991303,
|
|
"num_tokens": 9558578.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.3713301658630375,
|
|
"epoch": 4.196274420717856,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003695834305212925,
|
|
"loss": 4.9705,
|
|
"mean_token_accuracy": 0.19391765743494033,
|
|
"num_tokens": 9568538.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.416393518447876,
|
|
"epoch": 4.200817810086324,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.00036926262915463294,
|
|
"loss": 5.0126,
|
|
"mean_token_accuracy": 0.18483759909868241,
|
|
"num_tokens": 9578819.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.341962385177612,
|
|
"epoch": 4.205361199454793,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00036894159520430616,
|
|
"loss": 4.9418,
|
|
"mean_token_accuracy": 0.190701225399971,
|
|
"num_tokens": 9589060.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.397114658355713,
|
|
"epoch": 4.209904588823262,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00036862032946243173,
|
|
"loss": 5.0263,
|
|
"mean_token_accuracy": 0.18933148086071014,
|
|
"num_tokens": 9599332.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.4418495178222654,
|
|
"epoch": 4.214447978191731,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00036829883272170087,
|
|
"loss": 5.0274,
|
|
"mean_token_accuracy": 0.19345397502183914,
|
|
"num_tokens": 9610429.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.399268865585327,
|
|
"epoch": 4.2189913675602,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00036797710577537486,
|
|
"loss": 4.9947,
|
|
"mean_token_accuracy": 0.19395281225442887,
|
|
"num_tokens": 9620187.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.305547618865967,
|
|
"epoch": 4.223534756928669,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00036765514941728295,
|
|
"loss": 4.8942,
|
|
"mean_token_accuracy": 0.1970635548233986,
|
|
"num_tokens": 9629322.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.360532569885254,
|
|
"epoch": 4.228078146297138,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003673329644418204,
|
|
"loss": 5.0109,
|
|
"mean_token_accuracy": 0.19322539716959,
|
|
"num_tokens": 9639157.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.326487731933594,
|
|
"epoch": 4.232621535665607,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036701055164394664,
|
|
"loss": 5.008,
|
|
"mean_token_accuracy": 0.18961136937141418,
|
|
"num_tokens": 9649382.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.417474222183228,
|
|
"epoch": 4.237164925034075,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00036668791181918313,
|
|
"loss": 5.0017,
|
|
"mean_token_accuracy": 0.19651895761489868,
|
|
"num_tokens": 9659239.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.483715295791626,
|
|
"epoch": 4.241708314402544,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003663650457636117,
|
|
"loss": 5.0972,
|
|
"mean_token_accuracy": 0.18615776896476746,
|
|
"num_tokens": 9669308.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.349047422409058,
|
|
"epoch": 4.246251703771013,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003660419542738722,
|
|
"loss": 4.9972,
|
|
"mean_token_accuracy": 0.18265134543180467,
|
|
"num_tokens": 9679825.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.383522272109985,
|
|
"epoch": 4.250795093139482,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003657186381471608,
|
|
"loss": 5.0135,
|
|
"mean_token_accuracy": 0.18740416318178177,
|
|
"num_tokens": 9689709.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.374665451049805,
|
|
"epoch": 4.255338482507951,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00036539509818122765,
|
|
"loss": 4.9895,
|
|
"mean_token_accuracy": 0.19203248322010041,
|
|
"num_tokens": 9699186.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.334535360336304,
|
|
"epoch": 4.25988187187642,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003650713351743758,
|
|
"loss": 4.941,
|
|
"mean_token_accuracy": 0.19431956559419633,
|
|
"num_tokens": 9709967.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.4090399742126465,
|
|
"epoch": 4.264425261244889,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003647473499254581,
|
|
"loss": 5.1009,
|
|
"mean_token_accuracy": 0.17849036902189255,
|
|
"num_tokens": 9720604.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.438356685638428,
|
|
"epoch": 4.268968650613358,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000364423143233876,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.18394971489906312,
|
|
"num_tokens": 9731962.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.435949850082397,
|
|
"epoch": 4.273512039981826,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036409871589957723,
|
|
"loss": 5.0099,
|
|
"mean_token_accuracy": 0.19515601992607118,
|
|
"num_tokens": 9742254.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.439377737045288,
|
|
"epoch": 4.278055429350295,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000363774068723054,
|
|
"loss": 5.0837,
|
|
"mean_token_accuracy": 0.18270856142044067,
|
|
"num_tokens": 9753166.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.307160377502441,
|
|
"epoch": 4.282598818718764,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00036344920250534106,
|
|
"loss": 4.9764,
|
|
"mean_token_accuracy": 0.19560126066207886,
|
|
"num_tokens": 9764595.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.417747831344604,
|
|
"epoch": 4.287142208087233,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003631241180480133,
|
|
"loss": 5.075,
|
|
"mean_token_accuracy": 0.18836642503738404,
|
|
"num_tokens": 9775829.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.325015878677368,
|
|
"epoch": 4.291685597455702,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003627988161531845,
|
|
"loss": 4.9309,
|
|
"mean_token_accuracy": 0.1925303518772125,
|
|
"num_tokens": 9786068.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.3520349025726315,
|
|
"epoch": 4.296228986824171,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003624732976235046,
|
|
"loss": 5.0132,
|
|
"mean_token_accuracy": 0.1918956980109215,
|
|
"num_tokens": 9795757.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.347891473770142,
|
|
"epoch": 4.30077237619264,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003621475632621584,
|
|
"loss": 5.0528,
|
|
"mean_token_accuracy": 0.1867263376712799,
|
|
"num_tokens": 9806232.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.353366994857788,
|
|
"epoch": 4.305315765561109,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003618216138728627,
|
|
"loss": 5.0175,
|
|
"mean_token_accuracy": 0.19170364439487458,
|
|
"num_tokens": 9816251.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.421222496032715,
|
|
"epoch": 4.309859154929577,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00036149545025986567,
|
|
"loss": 4.9144,
|
|
"mean_token_accuracy": 0.19924249798059462,
|
|
"num_tokens": 9825935.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.3274205207824705,
|
|
"epoch": 4.314402544298046,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036116907322794335,
|
|
"loss": 5.0791,
|
|
"mean_token_accuracy": 0.1840093657374382,
|
|
"num_tokens": 9836575.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.413946151733398,
|
|
"epoch": 4.318945933666515,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00036084248358239857,
|
|
"loss": 5.0377,
|
|
"mean_token_accuracy": 0.18998847901821136,
|
|
"num_tokens": 9846872.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.374257230758667,
|
|
"epoch": 4.323489323034984,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036051568212905897,
|
|
"loss": 5.0052,
|
|
"mean_token_accuracy": 0.18385844975709914,
|
|
"num_tokens": 9858170.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.304642581939698,
|
|
"epoch": 4.328032712403453,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003601886696742745,
|
|
"loss": 4.9666,
|
|
"mean_token_accuracy": 0.19177217036485672,
|
|
"num_tokens": 9867497.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.311470079421997,
|
|
"epoch": 4.332576101771922,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000359861447024916,
|
|
"loss": 4.926,
|
|
"mean_token_accuracy": 0.18775672614574432,
|
|
"num_tokens": 9878749.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.419491386413574,
|
|
"epoch": 4.337119491140391,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00035953401498837285,
|
|
"loss": 4.9217,
|
|
"mean_token_accuracy": 0.20079722106456757,
|
|
"num_tokens": 9888438.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.3242302417755125,
|
|
"epoch": 4.34166288050886,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00035920637437255096,
|
|
"loss": 5.0508,
|
|
"mean_token_accuracy": 0.1838357925415039,
|
|
"num_tokens": 9898657.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.305213594436646,
|
|
"epoch": 4.346206269877328,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000358878525985871,
|
|
"loss": 5.0165,
|
|
"mean_token_accuracy": 0.19858954548835756,
|
|
"num_tokens": 9909012.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.406889200210571,
|
|
"epoch": 4.350749659245797,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003585504706372663,
|
|
"loss": 5.0446,
|
|
"mean_token_accuracy": 0.19165341258049012,
|
|
"num_tokens": 9919515.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.325256490707398,
|
|
"epoch": 4.355293048614266,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003582222091361808,
|
|
"loss": 4.8842,
|
|
"mean_token_accuracy": 0.20152657181024553,
|
|
"num_tokens": 9928877.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.286517715454101,
|
|
"epoch": 4.359836437982735,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003578937422925671,
|
|
"loss": 4.9648,
|
|
"mean_token_accuracy": 0.19535529166460036,
|
|
"num_tokens": 9938344.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.3873223781585695,
|
|
"epoch": 4.364379827351204,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003575650709168846,
|
|
"loss": 5.0685,
|
|
"mean_token_accuracy": 0.18586100190877913,
|
|
"num_tokens": 9949541.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.370128154754639,
|
|
"epoch": 4.368923216719673,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035723619582009715,
|
|
"loss": 5.0144,
|
|
"mean_token_accuracy": 0.18731602132320405,
|
|
"num_tokens": 9959894.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.337802124023438,
|
|
"epoch": 4.373466606088142,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00035690711781367137,
|
|
"loss": 4.9751,
|
|
"mean_token_accuracy": 0.19134166985750198,
|
|
"num_tokens": 9970291.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.399457502365112,
|
|
"epoch": 4.378009995456611,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003565778377095744,
|
|
"loss": 5.0666,
|
|
"mean_token_accuracy": 0.18581253290176392,
|
|
"num_tokens": 9981581.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.344257926940918,
|
|
"epoch": 4.38255338482508,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003562483563202724,
|
|
"loss": 4.9029,
|
|
"mean_token_accuracy": 0.1919662907719612,
|
|
"num_tokens": 9991472.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.313810205459594,
|
|
"epoch": 4.387096774193548,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00035591867445872784,
|
|
"loss": 4.98,
|
|
"mean_token_accuracy": 0.19492995142936706,
|
|
"num_tokens": 10002682.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.38421483039856,
|
|
"epoch": 4.391640163562017,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000355588792938398,
|
|
"loss": 5.0794,
|
|
"mean_token_accuracy": 0.18861068189144134,
|
|
"num_tokens": 10013856.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.339113712310791,
|
|
"epoch": 4.396183552930486,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003552587125732326,
|
|
"loss": 5.0198,
|
|
"mean_token_accuracy": 0.18613617718219758,
|
|
"num_tokens": 10024750.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.406014680862427,
|
|
"epoch": 4.400726942298955,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00035492843417767224,
|
|
"loss": 5.0342,
|
|
"mean_token_accuracy": 0.18661555051803588,
|
|
"num_tokens": 10034891.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.408040380477905,
|
|
"epoch": 4.405270331667424,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000354597958566646,
|
|
"loss": 4.9881,
|
|
"mean_token_accuracy": 0.19505923241376877,
|
|
"num_tokens": 10044860.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.2643787384033205,
|
|
"epoch": 4.409813721035893,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035426728655556954,
|
|
"loss": 4.9921,
|
|
"mean_token_accuracy": 0.19066058695316315,
|
|
"num_tokens": 10055959.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.381854820251465,
|
|
"epoch": 4.414357110404362,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003539364189603433,
|
|
"loss": 5.0751,
|
|
"mean_token_accuracy": 0.18194232285022735,
|
|
"num_tokens": 10067885.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.426373243331909,
|
|
"epoch": 4.418900499772831,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00035360535659735006,
|
|
"loss": 5.0815,
|
|
"mean_token_accuracy": 0.19203280955553054,
|
|
"num_tokens": 10078328.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.407176351547241,
|
|
"epoch": 4.423443889141299,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003532741002834534,
|
|
"loss": 4.9505,
|
|
"mean_token_accuracy": 0.1866741582751274,
|
|
"num_tokens": 10087880.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.33120493888855,
|
|
"epoch": 4.427987278509768,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00035294265083599526,
|
|
"loss": 5.0251,
|
|
"mean_token_accuracy": 0.18502462655305862,
|
|
"num_tokens": 10098432.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.313825464248657,
|
|
"epoch": 4.432530667878237,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003526110090727942,
|
|
"loss": 4.9801,
|
|
"mean_token_accuracy": 0.19674742221832275,
|
|
"num_tokens": 10108169.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.404324626922607,
|
|
"epoch": 4.437074057246706,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003522791758121434,
|
|
"loss": 5.0281,
|
|
"mean_token_accuracy": 0.18677810281515123,
|
|
"num_tokens": 10118580.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.287456274032593,
|
|
"epoch": 4.441617446615175,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00035194715187280835,
|
|
"loss": 4.8921,
|
|
"mean_token_accuracy": 0.19995710104703904,
|
|
"num_tokens": 10128259.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.339843368530273,
|
|
"epoch": 4.446160835983644,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003516149380740252,
|
|
"loss": 5.0035,
|
|
"mean_token_accuracy": 0.1915807768702507,
|
|
"num_tokens": 10137739.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.409734106063842,
|
|
"epoch": 4.450704225352113,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003512825352354983,
|
|
"loss": 5.0522,
|
|
"mean_token_accuracy": 0.1865141361951828,
|
|
"num_tokens": 10148886.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.350930643081665,
|
|
"epoch": 4.455247614720582,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00035094994417739884,
|
|
"loss": 4.9616,
|
|
"mean_token_accuracy": 0.19172403514385222,
|
|
"num_tokens": 10158680.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.276062965393066,
|
|
"epoch": 4.45979100408905,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000350617165720362,
|
|
"loss": 4.95,
|
|
"mean_token_accuracy": 0.1916424185037613,
|
|
"num_tokens": 10168893.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.2912571907043455,
|
|
"epoch": 4.464334393457519,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035028420068548567,
|
|
"loss": 4.8973,
|
|
"mean_token_accuracy": 0.19879554361104965,
|
|
"num_tokens": 10178781.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.3269970417022705,
|
|
"epoch": 4.468877782825988,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003499510498943279,
|
|
"loss": 5.018,
|
|
"mean_token_accuracy": 0.1866466745734215,
|
|
"num_tokens": 10189233.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.333923959732056,
|
|
"epoch": 4.473421172194457,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034961771416890526,
|
|
"loss": 4.9449,
|
|
"mean_token_accuracy": 0.19793631285429,
|
|
"num_tokens": 10198374.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.473109388351441,
|
|
"epoch": 4.477964561562926,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003492841943316903,
|
|
"loss": 5.0644,
|
|
"mean_token_accuracy": 0.18521908521652222,
|
|
"num_tokens": 10208924.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.376391410827637,
|
|
"epoch": 4.482507950931395,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003489504912056103,
|
|
"loss": 5.1419,
|
|
"mean_token_accuracy": 0.17972823977470398,
|
|
"num_tokens": 10219801.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.319911909103394,
|
|
"epoch": 4.487051340299864,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.0003486166056140445,
|
|
"loss": 4.9645,
|
|
"mean_token_accuracy": 0.18825983107089997,
|
|
"num_tokens": 10230233.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.38299970626831,
|
|
"epoch": 4.491594729668333,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034828253838082223,
|
|
"loss": 5.0863,
|
|
"mean_token_accuracy": 0.1837570071220398,
|
|
"num_tokens": 10241448.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.3804769039154055,
|
|
"epoch": 4.496138119036801,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034794829033022134,
|
|
"loss": 4.9434,
|
|
"mean_token_accuracy": 0.19734436869621277,
|
|
"num_tokens": 10250619.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.393267965316772,
|
|
"epoch": 4.50068150840527,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003476138622869657,
|
|
"loss": 5.06,
|
|
"mean_token_accuracy": 0.18689784407615662,
|
|
"num_tokens": 10260246.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.258755493164062,
|
|
"epoch": 4.505224897773739,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003472792550762232,
|
|
"loss": 4.9031,
|
|
"mean_token_accuracy": 0.194901442527771,
|
|
"num_tokens": 10269626.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.35222053527832,
|
|
"epoch": 4.509768287142208,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003469444695236038,
|
|
"loss": 5.0026,
|
|
"mean_token_accuracy": 0.19256832748651503,
|
|
"num_tokens": 10279954.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.301364088058472,
|
|
"epoch": 4.514311676510677,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00034660950645515765,
|
|
"loss": 4.9603,
|
|
"mean_token_accuracy": 0.1941036343574524,
|
|
"num_tokens": 10290099.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.359032869338989,
|
|
"epoch": 4.518855065879146,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003462743666973728,
|
|
"loss": 4.9505,
|
|
"mean_token_accuracy": 0.1945420503616333,
|
|
"num_tokens": 10299469.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.354590272903442,
|
|
"epoch": 4.523398455247615,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003459390510771734,
|
|
"loss": 4.927,
|
|
"mean_token_accuracy": 0.1967943012714386,
|
|
"num_tokens": 10310058.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.230786561965942,
|
|
"epoch": 4.527941844616084,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003456035604219172,
|
|
"loss": 4.9548,
|
|
"mean_token_accuracy": 0.19586427509784698,
|
|
"num_tokens": 10320960.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.263769149780273,
|
|
"epoch": 4.532485233984552,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003452678955593942,
|
|
"loss": 4.9948,
|
|
"mean_token_accuracy": 0.1886756971478462,
|
|
"num_tokens": 10331931.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.344566392898559,
|
|
"epoch": 4.537028623353021,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000344932057317824,
|
|
"loss": 5.0327,
|
|
"mean_token_accuracy": 0.18660142719745637,
|
|
"num_tokens": 10341964.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.381228446960449,
|
|
"epoch": 4.54157201272149,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003445960465258542,
|
|
"loss": 5.1099,
|
|
"mean_token_accuracy": 0.1798148900270462,
|
|
"num_tokens": 10352456.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.54157201272149,
|
|
"eval_entropy": 5.295619406811027,
|
|
"eval_loss": 5.797823429107666,
|
|
"eval_mean_token_accuracy": 0.1628394648606001,
|
|
"eval_num_tokens": 10352456.0,
|
|
"eval_runtime": 2.5968,
|
|
"eval_samples_per_second": 1322.792,
|
|
"eval_steps_per_second": 165.59,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.379141473770142,
|
|
"epoch": 4.546115402089959,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003442598640125578,
|
|
"loss": 5.0007,
|
|
"mean_token_accuracy": 0.1900293231010437,
|
|
"num_tokens": 10363672.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.361876058578491,
|
|
"epoch": 4.550658791458428,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000343923510607432,
|
|
"loss": 5.0898,
|
|
"mean_token_accuracy": 0.18669371157884598,
|
|
"num_tokens": 10374602.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.276431751251221,
|
|
"epoch": 4.555202180826897,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003435869871403952,
|
|
"loss": 4.9284,
|
|
"mean_token_accuracy": 0.19740553498268126,
|
|
"num_tokens": 10384374.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.373429250717163,
|
|
"epoch": 4.559745570195366,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034325029444178565,
|
|
"loss": 5.1057,
|
|
"mean_token_accuracy": 0.18404812961816788,
|
|
"num_tokens": 10395949.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.377804470062256,
|
|
"epoch": 4.564288959563835,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034291343334235915,
|
|
"loss": 5.0251,
|
|
"mean_token_accuracy": 0.19063410311937332,
|
|
"num_tokens": 10407162.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.298324298858643,
|
|
"epoch": 4.568832348932304,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003425764046732869,
|
|
"loss": 4.9806,
|
|
"mean_token_accuracy": 0.18880284279584886,
|
|
"num_tokens": 10418191.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.4157843589782715,
|
|
"epoch": 4.573375738300772,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00034223920926615376,
|
|
"loss": 5.0181,
|
|
"mean_token_accuracy": 0.19342149496078492,
|
|
"num_tokens": 10427899.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.305647468566894,
|
|
"epoch": 4.577919127669241,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00034190184795295577,
|
|
"loss": 4.9661,
|
|
"mean_token_accuracy": 0.19671805649995805,
|
|
"num_tokens": 10437882.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.269737195968628,
|
|
"epoch": 4.58246251703771,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 0.00034156432156609844,
|
|
"loss": 4.9882,
|
|
"mean_token_accuracy": 0.19156125634908677,
|
|
"num_tokens": 10448111.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.332058048248291,
|
|
"epoch": 4.587005906406179,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00034122663093839466,
|
|
"loss": 4.9762,
|
|
"mean_token_accuracy": 0.19052882939577104,
|
|
"num_tokens": 10457522.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.345019769668579,
|
|
"epoch": 4.591549295774648,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.0003408887769030624,
|
|
"loss": 4.9035,
|
|
"mean_token_accuracy": 0.20072456151247026,
|
|
"num_tokens": 10467229.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.293692874908447,
|
|
"epoch": 4.596092685143117,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000340550760293723,
|
|
"loss": 4.9776,
|
|
"mean_token_accuracy": 0.19640641361474992,
|
|
"num_tokens": 10477504.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.343152761459351,
|
|
"epoch": 4.600636074511586,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00034021258194439876,
|
|
"loss": 4.9484,
|
|
"mean_token_accuracy": 0.19706695526838303,
|
|
"num_tokens": 10487238.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.3097553730010985,
|
|
"epoch": 4.605179463880054,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00033987424268951117,
|
|
"loss": 5.0247,
|
|
"mean_token_accuracy": 0.18263342678546907,
|
|
"num_tokens": 10497061.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.351561164855957,
|
|
"epoch": 4.609722853248523,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00033953574336387867,
|
|
"loss": 4.9995,
|
|
"mean_token_accuracy": 0.20171792507171632,
|
|
"num_tokens": 10507763.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.3355754852294925,
|
|
"epoch": 4.614266242616992,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003391970848027148,
|
|
"loss": 4.9959,
|
|
"mean_token_accuracy": 0.19700370579957963,
|
|
"num_tokens": 10518466.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.322419738769531,
|
|
"epoch": 4.618809631985461,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003388582678416257,
|
|
"loss": 5.0478,
|
|
"mean_token_accuracy": 0.1906829372048378,
|
|
"num_tokens": 10528125.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.321761274337769,
|
|
"epoch": 4.62335302135393,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00033851929331660876,
|
|
"loss": 4.9688,
|
|
"mean_token_accuracy": 0.19541918039321898,
|
|
"num_tokens": 10537684.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.37051043510437,
|
|
"epoch": 4.627896410722399,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033818016206404983,
|
|
"loss": 5.0696,
|
|
"mean_token_accuracy": 0.18494369089603424,
|
|
"num_tokens": 10548442.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.383481168746949,
|
|
"epoch": 4.632439800090868,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003378408749207215,
|
|
"loss": 5.0028,
|
|
"mean_token_accuracy": 0.19469185024499894,
|
|
"num_tokens": 10558885.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.293408298492432,
|
|
"epoch": 4.636983189459337,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003375014327237812,
|
|
"loss": 4.9504,
|
|
"mean_token_accuracy": 0.19533636271953583,
|
|
"num_tokens": 10570481.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.37161545753479,
|
|
"epoch": 4.641526578827806,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003371618363107686,
|
|
"loss": 5.0375,
|
|
"mean_token_accuracy": 0.19561344683170317,
|
|
"num_tokens": 10580042.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.313982677459717,
|
|
"epoch": 4.646069968196274,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00033682208651960444,
|
|
"loss": 5.02,
|
|
"mean_token_accuracy": 0.18770121186971664,
|
|
"num_tokens": 10590588.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.356268405914307,
|
|
"epoch": 4.650613357564743,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00033648218418858716,
|
|
"loss": 4.9881,
|
|
"mean_token_accuracy": 0.18482399582862855,
|
|
"num_tokens": 10600952.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.400371789932251,
|
|
"epoch": 4.655156746933212,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.0003361421301563922,
|
|
"loss": 5.0171,
|
|
"mean_token_accuracy": 0.19349045753479005,
|
|
"num_tokens": 10610447.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.358433771133423,
|
|
"epoch": 4.659700136301681,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003358019252620691,
|
|
"loss": 5.052,
|
|
"mean_token_accuracy": 0.18518519401550293,
|
|
"num_tokens": 10621314.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.397951555252075,
|
|
"epoch": 4.66424352567015,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033546157034503954,
|
|
"loss": 5.0807,
|
|
"mean_token_accuracy": 0.1906274989247322,
|
|
"num_tokens": 10631879.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.419374990463257,
|
|
"epoch": 4.668786915038619,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003351210662450955,
|
|
"loss": 4.9873,
|
|
"mean_token_accuracy": 0.19436534494161606,
|
|
"num_tokens": 10642233.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.362288761138916,
|
|
"epoch": 4.673330304407088,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0003347804138023968,
|
|
"loss": 5.0469,
|
|
"mean_token_accuracy": 0.19054364562034606,
|
|
"num_tokens": 10652869.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.307768678665161,
|
|
"epoch": 4.677873693775557,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003344396138574698,
|
|
"loss": 5.0056,
|
|
"mean_token_accuracy": 0.1876662254333496,
|
|
"num_tokens": 10663368.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.447361660003662,
|
|
"epoch": 4.682417083144026,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003340986672512042,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.18089387118816375,
|
|
"num_tokens": 10674574.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.503390121459961,
|
|
"epoch": 4.686960472512494,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003337575748248521,
|
|
"loss": 5.1809,
|
|
"mean_token_accuracy": 0.17412161827087402,
|
|
"num_tokens": 10685935.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.33093843460083,
|
|
"epoch": 4.691503861880963,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00033341633742002503,
|
|
"loss": 5.0522,
|
|
"mean_token_accuracy": 0.18769600987434387,
|
|
"num_tokens": 10696054.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.335041809082031,
|
|
"epoch": 4.696047251249432,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003330749558786924,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.1874743789434433,
|
|
"num_tokens": 10706334.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.398832559585571,
|
|
"epoch": 4.700590640617901,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003327334310431791,
|
|
"loss": 5.0067,
|
|
"mean_token_accuracy": 0.19138273447752,
|
|
"num_tokens": 10716953.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.338293504714966,
|
|
"epoch": 4.70513402998637,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003323917637561638,
|
|
"loss": 4.9986,
|
|
"mean_token_accuracy": 0.19358458667993544,
|
|
"num_tokens": 10726633.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.262207698822022,
|
|
"epoch": 4.709677419354839,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033204995486067664,
|
|
"loss": 4.9402,
|
|
"mean_token_accuracy": 0.19110361635684966,
|
|
"num_tokens": 10735862.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.333451271057129,
|
|
"epoch": 4.714220808723308,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033170800520009703,
|
|
"loss": 5.0124,
|
|
"mean_token_accuracy": 0.1893896386027336,
|
|
"num_tokens": 10746505.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.3548023223876955,
|
|
"epoch": 4.718764198091776,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003313659156181518,
|
|
"loss": 4.9437,
|
|
"mean_token_accuracy": 0.19869993925094603,
|
|
"num_tokens": 10756513.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.341909790039063,
|
|
"epoch": 4.723307587460245,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033102368695891283,
|
|
"loss": 4.9909,
|
|
"mean_token_accuracy": 0.1934805154800415,
|
|
"num_tokens": 10766855.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.335283184051514,
|
|
"epoch": 4.727850976828714,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00033068132006679554,
|
|
"loss": 4.9527,
|
|
"mean_token_accuracy": 0.19475856870412828,
|
|
"num_tokens": 10776213.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.368388557434082,
|
|
"epoch": 4.732394366197183,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00033033881578655605,
|
|
"loss": 5.098,
|
|
"mean_token_accuracy": 0.18702502846717833,
|
|
"num_tokens": 10786476.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.3042230129241945,
|
|
"epoch": 4.736937755565652,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003299961749632898,
|
|
"loss": 4.8958,
|
|
"mean_token_accuracy": 0.19572898000478745,
|
|
"num_tokens": 10796274.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.337723064422607,
|
|
"epoch": 4.741481144934121,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032965339844242864,
|
|
"loss": 5.0601,
|
|
"mean_token_accuracy": 0.19185242354869841,
|
|
"num_tokens": 10806337.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.29773964881897,
|
|
"epoch": 4.74602453430259,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003293104870697399,
|
|
"loss": 4.9385,
|
|
"mean_token_accuracy": 0.19391794353723527,
|
|
"num_tokens": 10816403.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.328159189224243,
|
|
"epoch": 4.750567923671059,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032896744169132305,
|
|
"loss": 5.1111,
|
|
"mean_token_accuracy": 0.18056605458259584,
|
|
"num_tokens": 10827830.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.395383882522583,
|
|
"epoch": 4.755111313039528,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003286242631536087,
|
|
"loss": 5.0511,
|
|
"mean_token_accuracy": 0.1874497964978218,
|
|
"num_tokens": 10839057.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.390955114364624,
|
|
"epoch": 4.759654702407996,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00032828095230335555,
|
|
"loss": 5.0932,
|
|
"mean_token_accuracy": 0.1828780487179756,
|
|
"num_tokens": 10849814.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.348363161087036,
|
|
"epoch": 4.764198091776465,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003279375099876491,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.1881968930363655,
|
|
"num_tokens": 10860848.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.354051017761231,
|
|
"epoch": 4.768741481144934,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003275939370538992,
|
|
"loss": 4.9969,
|
|
"mean_token_accuracy": 0.19064449071884154,
|
|
"num_tokens": 10870349.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.392513227462769,
|
|
"epoch": 4.773284870513403,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032725023434983767,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.1887713223695755,
|
|
"num_tokens": 10880519.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.374751710891724,
|
|
"epoch": 4.777828259881872,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00032690640272351715,
|
|
"loss": 4.9887,
|
|
"mean_token_accuracy": 0.19271645843982696,
|
|
"num_tokens": 10890254.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.303708553314209,
|
|
"epoch": 4.782371649250341,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003265624430233076,
|
|
"loss": 5.042,
|
|
"mean_token_accuracy": 0.18023288398981094,
|
|
"num_tokens": 10901127.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.316749334335327,
|
|
"epoch": 4.78691503861881,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003262183560978956,
|
|
"loss": 4.9831,
|
|
"mean_token_accuracy": 0.1916854664683342,
|
|
"num_tokens": 10911572.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.309233093261719,
|
|
"epoch": 4.791458427987278,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003258741427962814,
|
|
"loss": 4.9415,
|
|
"mean_token_accuracy": 0.19793882220983505,
|
|
"num_tokens": 10922518.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.377589702606201,
|
|
"epoch": 4.796001817355747,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00032552980396777714,
|
|
"loss": 5.0399,
|
|
"mean_token_accuracy": 0.18733723908662797,
|
|
"num_tokens": 10932818.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.4046424388885494,
|
|
"epoch": 4.800545206724216,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00032518534046200464,
|
|
"loss": 5.0585,
|
|
"mean_token_accuracy": 0.18666841089725494,
|
|
"num_tokens": 10943311.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.3893482208251955,
|
|
"epoch": 4.805088596092685,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0003248407531288932,
|
|
"loss": 5.0637,
|
|
"mean_token_accuracy": 0.18230961710214616,
|
|
"num_tokens": 10954157.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.3508580207824705,
|
|
"epoch": 4.809631985461154,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000324496042818678,
|
|
"loss": 5.0202,
|
|
"mean_token_accuracy": 0.19039032757282257,
|
|
"num_tokens": 10964745.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.361133861541748,
|
|
"epoch": 4.814175374829623,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00032415121038189746,
|
|
"loss": 5.0099,
|
|
"mean_token_accuracy": 0.1938726171851158,
|
|
"num_tokens": 10975123.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.392160654067993,
|
|
"epoch": 4.818718764198092,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00032380625666939125,
|
|
"loss": 5.0017,
|
|
"mean_token_accuracy": 0.1926610291004181,
|
|
"num_tokens": 10985189.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.28259015083313,
|
|
"epoch": 4.823262153566561,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00032346118253229834,
|
|
"loss": 4.9709,
|
|
"mean_token_accuracy": 0.18847636133432388,
|
|
"num_tokens": 10994990.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.280781173706055,
|
|
"epoch": 4.82780554293503,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00032311598882205496,
|
|
"loss": 4.9456,
|
|
"mean_token_accuracy": 0.20355015099048615,
|
|
"num_tokens": 11004989.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.389767360687256,
|
|
"epoch": 4.832348932303498,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00032277067639039216,
|
|
"loss": 5.0097,
|
|
"mean_token_accuracy": 0.19363986253738402,
|
|
"num_tokens": 11015225.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.413976240158081,
|
|
"epoch": 4.836892321671967,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00032242524608933423,
|
|
"loss": 5.042,
|
|
"mean_token_accuracy": 0.19259006083011626,
|
|
"num_tokens": 11026308.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.337032175064087,
|
|
"epoch": 4.841435711040436,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00032207969877119587,
|
|
"loss": 5.0021,
|
|
"mean_token_accuracy": 0.18526588529348373,
|
|
"num_tokens": 11036313.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.303933382034302,
|
|
"epoch": 4.845979100408905,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00032173403528858093,
|
|
"loss": 5.0051,
|
|
"mean_token_accuracy": 0.19635174572467803,
|
|
"num_tokens": 11046444.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.448164939880371,
|
|
"epoch": 4.850522489777374,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003213882564943796,
|
|
"loss": 5.0094,
|
|
"mean_token_accuracy": 0.19432779848575593,
|
|
"num_tokens": 11056514.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.352915716171265,
|
|
"epoch": 4.855065879145843,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003210423632417666,
|
|
"loss": 4.9772,
|
|
"mean_token_accuracy": 0.1921101316809654,
|
|
"num_tokens": 11066807.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.302493572235107,
|
|
"epoch": 4.859609268514312,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003206963563841993,
|
|
"loss": 5.0155,
|
|
"mean_token_accuracy": 0.18874137550592424,
|
|
"num_tokens": 11076141.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.325728511810302,
|
|
"epoch": 4.86415265788278,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032035023677541515,
|
|
"loss": 4.8999,
|
|
"mean_token_accuracy": 0.2012683257460594,
|
|
"num_tokens": 11085684.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.266070938110351,
|
|
"epoch": 4.868696047251249,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003200040052694299,
|
|
"loss": 4.9027,
|
|
"mean_token_accuracy": 0.1965961903333664,
|
|
"num_tokens": 11095401.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.330615711212158,
|
|
"epoch": 4.873239436619718,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00031965766272053534,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.187218376994133,
|
|
"num_tokens": 11105507.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.367094039916992,
|
|
"epoch": 4.877782825988187,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00031931120998329737,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.19982756227254866,
|
|
"num_tokens": 11114809.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.377289533615112,
|
|
"epoch": 4.882326215356656,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003189646479125536,
|
|
"loss": 5.0921,
|
|
"mean_token_accuracy": 0.18817085027694702,
|
|
"num_tokens": 11124870.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.410494422912597,
|
|
"epoch": 4.886869604725125,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003186179773634115,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.186170357465744,
|
|
"num_tokens": 11135925.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.337065410614014,
|
|
"epoch": 4.891412994093594,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00031827119919124614,
|
|
"loss": 4.9918,
|
|
"mean_token_accuracy": 0.19234817773103713,
|
|
"num_tokens": 11146494.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.319728088378906,
|
|
"epoch": 4.895956383462063,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00031792431425169826,
|
|
"loss": 4.9825,
|
|
"mean_token_accuracy": 0.19589998722076415,
|
|
"num_tokens": 11155817.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.352860736846924,
|
|
"epoch": 4.900499772830532,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00031757732340067195,
|
|
"loss": 5.0984,
|
|
"mean_token_accuracy": 0.18309517353773117,
|
|
"num_tokens": 11167129.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.414812469482422,
|
|
"epoch": 4.905043162199,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00031723022749433267,
|
|
"loss": 5.0729,
|
|
"mean_token_accuracy": 0.18558370918035508,
|
|
"num_tokens": 11177115.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.387073755264282,
|
|
"epoch": 4.909586551567469,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000316883027389105,
|
|
"loss": 5.0323,
|
|
"mean_token_accuracy": 0.1813766673207283,
|
|
"num_tokens": 11188551.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.36614875793457,
|
|
"epoch": 4.914129940935938,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003165357239416706,
|
|
"loss": 5.081,
|
|
"mean_token_accuracy": 0.18132559359073638,
|
|
"num_tokens": 11199560.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.383052062988281,
|
|
"epoch": 4.918673330304407,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003161883180089665,
|
|
"loss": 4.9805,
|
|
"mean_token_accuracy": 0.1930908367037773,
|
|
"num_tokens": 11210265.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.3928216934204105,
|
|
"epoch": 4.923216719672876,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003158408104481821,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.1843537360429764,
|
|
"num_tokens": 11221649.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.283775568008423,
|
|
"epoch": 4.927760109041345,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00031549320211675776,
|
|
"loss": 4.9331,
|
|
"mean_token_accuracy": 0.19778723418712615,
|
|
"num_tokens": 11232417.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.326270723342896,
|
|
"epoch": 4.932303498409814,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00031514549387238246,
|
|
"loss": 5.0632,
|
|
"mean_token_accuracy": 0.1845440998673439,
|
|
"num_tokens": 11243022.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.372826623916626,
|
|
"epoch": 4.936846887778283,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003147976865729918,
|
|
"loss": 5.0119,
|
|
"mean_token_accuracy": 0.1921074166893959,
|
|
"num_tokens": 11253663.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.372944593429565,
|
|
"epoch": 4.941390277146752,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00031444978107676567,
|
|
"loss": 5.0727,
|
|
"mean_token_accuracy": 0.1882585272192955,
|
|
"num_tokens": 11263482.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.345783996582031,
|
|
"epoch": 4.94593366651522,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00031410177824212635,
|
|
"loss": 4.9419,
|
|
"mean_token_accuracy": 0.19171193689107896,
|
|
"num_tokens": 11273956.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.334301853179932,
|
|
"epoch": 4.950477055883689,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003137536789277362,
|
|
"loss": 5.0689,
|
|
"mean_token_accuracy": 0.18793885856866838,
|
|
"num_tokens": 11283590.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.3658356189727785,
|
|
"epoch": 4.955020445252158,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003134054839924957,
|
|
"loss": 5.01,
|
|
"mean_token_accuracy": 0.18835936337709427,
|
|
"num_tokens": 11294724.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.393179225921631,
|
|
"epoch": 4.959563834620627,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00031305719429554124,
|
|
"loss": 4.9726,
|
|
"mean_token_accuracy": 0.19251521676778793,
|
|
"num_tokens": 11305995.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.293840646743774,
|
|
"epoch": 4.964107223989096,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003127088106962429,
|
|
"loss": 4.9583,
|
|
"mean_token_accuracy": 0.19573341310024261,
|
|
"num_tokens": 11315373.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.300927782058716,
|
|
"epoch": 4.968650613357565,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00031236033405420284,
|
|
"loss": 4.9718,
|
|
"mean_token_accuracy": 0.18980509489774705,
|
|
"num_tokens": 11325662.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.4405519485473635,
|
|
"epoch": 4.973194002726034,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00031201176522925236,
|
|
"loss": 5.157,
|
|
"mean_token_accuracy": 0.17674886733293532,
|
|
"num_tokens": 11335689.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.404199028015137,
|
|
"epoch": 4.977737392094502,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003116631050814503,
|
|
"loss": 4.9989,
|
|
"mean_token_accuracy": 0.19951657950878143,
|
|
"num_tokens": 11345501.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.309408044815063,
|
|
"epoch": 4.982280781462971,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031131435447108116,
|
|
"loss": 4.9506,
|
|
"mean_token_accuracy": 0.19686514884233475,
|
|
"num_tokens": 11356146.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.331334018707276,
|
|
"epoch": 4.98682417083144,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003109655142586521,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.1998109072446823,
|
|
"num_tokens": 11366507.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.360395431518555,
|
|
"epoch": 4.991367560199909,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00031061658530489195,
|
|
"loss": 5.0651,
|
|
"mean_token_accuracy": 0.18294547498226166,
|
|
"num_tokens": 11376479.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.364800977706909,
|
|
"epoch": 4.995910949568378,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00031026756847074804,
|
|
"loss": 5.0891,
|
|
"mean_token_accuracy": 0.18100034594535827,
|
|
"num_tokens": 11387863.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.995910949568378,
|
|
"eval_entropy": 5.346990779388783,
|
|
"eval_loss": 5.7602643966674805,
|
|
"eval_mean_token_accuracy": 0.16645758994096935,
|
|
"eval_num_tokens": 11387863.0,
|
|
"eval_runtime": 2.5765,
|
|
"eval_samples_per_second": 1333.185,
|
|
"eval_steps_per_second": 166.891,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.3882790671454535,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.0003099184646173849,
|
|
"loss": 4.9419,
|
|
"mean_token_accuracy": 0.19703455766042074,
|
|
"num_tokens": 11396985.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.323262500762939,
|
|
"epoch": 5.004543389368469,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.00030956927460618126,
|
|
"loss": 4.8201,
|
|
"mean_token_accuracy": 0.2017674505710602,
|
|
"num_tokens": 11406519.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.349852466583252,
|
|
"epoch": 5.009086778736938,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003092199992987291,
|
|
"loss": 4.9252,
|
|
"mean_token_accuracy": 0.19987574219703674,
|
|
"num_tokens": 11416746.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.335919904708862,
|
|
"epoch": 5.013630168105407,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00030887063955683044,
|
|
"loss": 4.7802,
|
|
"mean_token_accuracy": 0.20374078601598739,
|
|
"num_tokens": 11427218.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.308040428161621,
|
|
"epoch": 5.018173557473876,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00030852119624249553,
|
|
"loss": 4.7654,
|
|
"mean_token_accuracy": 0.20291833877563475,
|
|
"num_tokens": 11437623.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.353977346420288,
|
|
"epoch": 5.022716946842344,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003081716702179412,
|
|
"loss": 4.7032,
|
|
"mean_token_accuracy": 0.20726612359285354,
|
|
"num_tokens": 11446438.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.339612627029419,
|
|
"epoch": 5.027260336210813,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000307822062345588,
|
|
"loss": 4.8589,
|
|
"mean_token_accuracy": 0.1977094054222107,
|
|
"num_tokens": 11457401.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.286742210388184,
|
|
"epoch": 5.031803725579282,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0003074723734880588,
|
|
"loss": 4.8038,
|
|
"mean_token_accuracy": 0.20265266597270964,
|
|
"num_tokens": 11468295.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.3144087314605715,
|
|
"epoch": 5.036347114947751,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000307122604508176,
|
|
"loss": 4.761,
|
|
"mean_token_accuracy": 0.20199826806783677,
|
|
"num_tokens": 11478687.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.337947416305542,
|
|
"epoch": 5.04089050431622,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003067727562689598,
|
|
"loss": 4.8142,
|
|
"mean_token_accuracy": 0.19958952963352203,
|
|
"num_tokens": 11488121.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.309422492980957,
|
|
"epoch": 5.045433893684689,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003064228296336259,
|
|
"loss": 4.8205,
|
|
"mean_token_accuracy": 0.1989064931869507,
|
|
"num_tokens": 11498129.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.336434936523437,
|
|
"epoch": 5.049977283053158,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003060728254655836,
|
|
"loss": 4.8117,
|
|
"mean_token_accuracy": 0.1999408408999443,
|
|
"num_tokens": 11508870.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.360090684890747,
|
|
"epoch": 5.054520672421627,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00030572274462843327,
|
|
"loss": 4.8826,
|
|
"mean_token_accuracy": 0.20137253552675247,
|
|
"num_tokens": 11518333.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.299070739746094,
|
|
"epoch": 5.059064061790095,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030537258798596465,
|
|
"loss": 4.7989,
|
|
"mean_token_accuracy": 0.2009906217455864,
|
|
"num_tokens": 11529106.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.346049785614014,
|
|
"epoch": 5.063607451158564,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003050223564021543,
|
|
"loss": 4.9242,
|
|
"mean_token_accuracy": 0.1951611489057541,
|
|
"num_tokens": 11540248.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.307297086715698,
|
|
"epoch": 5.068150840527033,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003046720507411641,
|
|
"loss": 4.7568,
|
|
"mean_token_accuracy": 0.20703721195459365,
|
|
"num_tokens": 11550291.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.351901626586914,
|
|
"epoch": 5.072694229895502,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003043216718673383,
|
|
"loss": 4.8351,
|
|
"mean_token_accuracy": 0.20133900940418242,
|
|
"num_tokens": 11560915.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.311527395248413,
|
|
"epoch": 5.077237619263971,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00030397122064520186,
|
|
"loss": 4.8411,
|
|
"mean_token_accuracy": 0.19826627522706985,
|
|
"num_tokens": 11571169.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.37898850440979,
|
|
"epoch": 5.08178100863244,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00030362069793945856,
|
|
"loss": 4.9074,
|
|
"mean_token_accuracy": 0.19560447037220002,
|
|
"num_tokens": 11582784.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.4012579917907715,
|
|
"epoch": 5.086324398000909,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003032701046149881,
|
|
"loss": 4.8803,
|
|
"mean_token_accuracy": 0.2028103679418564,
|
|
"num_tokens": 11593160.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.289191865921021,
|
|
"epoch": 5.090867787369378,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0003029194415368448,
|
|
"loss": 4.7571,
|
|
"mean_token_accuracy": 0.20651391446590422,
|
|
"num_tokens": 11603277.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.308057832717895,
|
|
"epoch": 5.095411176737846,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0003025687095702548,
|
|
"loss": 4.8348,
|
|
"mean_token_accuracy": 0.20365611761808394,
|
|
"num_tokens": 11612847.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.2692125797271725,
|
|
"epoch": 5.099954566106315,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003022179095806146,
|
|
"loss": 4.8083,
|
|
"mean_token_accuracy": 0.20133823603391648,
|
|
"num_tokens": 11623090.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.322366571426391,
|
|
"epoch": 5.104497955474784,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000301867042433488,
|
|
"loss": 4.7558,
|
|
"mean_token_accuracy": 0.21492936909198762,
|
|
"num_tokens": 11634430.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.2926112651824955,
|
|
"epoch": 5.109041344843253,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00030151610899460505,
|
|
"loss": 4.8541,
|
|
"mean_token_accuracy": 0.2012924626469612,
|
|
"num_tokens": 11645425.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.301137971878052,
|
|
"epoch": 5.113584734211722,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00030116511012985916,
|
|
"loss": 4.8292,
|
|
"mean_token_accuracy": 0.20135027319192886,
|
|
"num_tokens": 11655446.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.356033372879028,
|
|
"epoch": 5.118128123580191,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003008140467053048,
|
|
"loss": 4.8624,
|
|
"mean_token_accuracy": 0.19674912095069885,
|
|
"num_tokens": 11665383.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.3330309867858885,
|
|
"epoch": 5.12267151294866,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0003004629195871565,
|
|
"loss": 4.7962,
|
|
"mean_token_accuracy": 0.20144290179014207,
|
|
"num_tokens": 11676366.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.285286235809326,
|
|
"epoch": 5.127214902317129,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003001117296417852,
|
|
"loss": 4.8138,
|
|
"mean_token_accuracy": 0.20153769850730896,
|
|
"num_tokens": 11686537.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.354802560806275,
|
|
"epoch": 5.131758291685597,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002997604777357173,
|
|
"loss": 4.9536,
|
|
"mean_token_accuracy": 0.19828756898641586,
|
|
"num_tokens": 11698142.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.319665575027466,
|
|
"epoch": 5.136301681054066,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000299409164735632,
|
|
"loss": 4.819,
|
|
"mean_token_accuracy": 0.20249372571706772,
|
|
"num_tokens": 11709457.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.377064514160156,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0002990577915083591,
|
|
"loss": 4.7941,
|
|
"mean_token_accuracy": 0.2089761346578598,
|
|
"num_tokens": 11719077.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.311314821243286,
|
|
"epoch": 5.145388459791004,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029870635892087716,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.2011446088552475,
|
|
"num_tokens": 11728823.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.327394294738769,
|
|
"epoch": 5.149931849159473,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002983548678403112,
|
|
"loss": 4.8281,
|
|
"mean_token_accuracy": 0.20240188986063004,
|
|
"num_tokens": 11739574.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.2840265274047855,
|
|
"epoch": 5.154475238527942,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002980033191339305,
|
|
"loss": 4.8132,
|
|
"mean_token_accuracy": 0.2048185423016548,
|
|
"num_tokens": 11749499.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.222664833068848,
|
|
"epoch": 5.159018627896411,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002976517136691465,
|
|
"loss": 4.7177,
|
|
"mean_token_accuracy": 0.2104017823934555,
|
|
"num_tokens": 11759438.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.223516845703125,
|
|
"epoch": 5.16356201726488,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00029730005231351074,
|
|
"loss": 4.7451,
|
|
"mean_token_accuracy": 0.2085548609495163,
|
|
"num_tokens": 11769432.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.321575021743774,
|
|
"epoch": 5.168105406633348,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002969483359347127,
|
|
"loss": 4.8236,
|
|
"mean_token_accuracy": 0.20101623237133026,
|
|
"num_tokens": 11780886.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.340514373779297,
|
|
"epoch": 5.172648796001817,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002965965654005776,
|
|
"loss": 4.8965,
|
|
"mean_token_accuracy": 0.20480832308530808,
|
|
"num_tokens": 11791824.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.241022682189941,
|
|
"epoch": 5.177192185370286,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002962447415790642,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.2063835322856903,
|
|
"num_tokens": 11801941.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.314211702346801,
|
|
"epoch": 5.181735574738755,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029589286533826276,
|
|
"loss": 4.8635,
|
|
"mean_token_accuracy": 0.199500373005867,
|
|
"num_tokens": 11812258.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.385710430145264,
|
|
"epoch": 5.186278964107224,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002955409375463931,
|
|
"loss": 4.9644,
|
|
"mean_token_accuracy": 0.19458112567663194,
|
|
"num_tokens": 11823046.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.386393928527832,
|
|
"epoch": 5.190822353475693,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002951889590718019,
|
|
"loss": 4.8824,
|
|
"mean_token_accuracy": 0.19672468900680543,
|
|
"num_tokens": 11834080.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.276735830307007,
|
|
"epoch": 5.195365742844162,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002948369307829612,
|
|
"loss": 4.7754,
|
|
"mean_token_accuracy": 0.21169690638780594,
|
|
"num_tokens": 11843848.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.218534994125366,
|
|
"epoch": 5.199909132212631,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002944848535484656,
|
|
"loss": 4.7438,
|
|
"mean_token_accuracy": 0.21150794476270676,
|
|
"num_tokens": 11854429.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.2952666759490965,
|
|
"epoch": 5.2044525215811,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029413272823703085,
|
|
"loss": 4.9137,
|
|
"mean_token_accuracy": 0.19688897877931594,
|
|
"num_tokens": 11865035.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.403980255126953,
|
|
"epoch": 5.208995910949568,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002937805557174911,
|
|
"loss": 4.8039,
|
|
"mean_token_accuracy": 0.2036522775888443,
|
|
"num_tokens": 11875829.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.362405109405517,
|
|
"epoch": 5.213539300318037,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00029342833685879713,
|
|
"loss": 4.8763,
|
|
"mean_token_accuracy": 0.1964132681488991,
|
|
"num_tokens": 11886397.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.342545080184936,
|
|
"epoch": 5.218082689686506,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002930760725300137,
|
|
"loss": 4.8805,
|
|
"mean_token_accuracy": 0.20535845011472703,
|
|
"num_tokens": 11896619.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.304865217208862,
|
|
"epoch": 5.222626079054975,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00029272376360031833,
|
|
"loss": 4.8861,
|
|
"mean_token_accuracy": 0.19610541313886642,
|
|
"num_tokens": 11906825.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.293488502502441,
|
|
"epoch": 5.227169468423444,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00029237141093899804,
|
|
"loss": 4.8367,
|
|
"mean_token_accuracy": 0.1997443750500679,
|
|
"num_tokens": 11917051.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.281969308853149,
|
|
"epoch": 5.231712857791913,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00029201901541544824,
|
|
"loss": 4.824,
|
|
"mean_token_accuracy": 0.2063928425312042,
|
|
"num_tokens": 11927830.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.3192009925842285,
|
|
"epoch": 5.236256247160382,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002916665778991696,
|
|
"loss": 4.7963,
|
|
"mean_token_accuracy": 0.2009850785136223,
|
|
"num_tokens": 11937541.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.3072185039520265,
|
|
"epoch": 5.240799636528851,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002913140992597668,
|
|
"loss": 4.8373,
|
|
"mean_token_accuracy": 0.2010096177458763,
|
|
"num_tokens": 11947915.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.320555877685547,
|
|
"epoch": 5.245343025897319,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002909615803669458,
|
|
"loss": 4.7635,
|
|
"mean_token_accuracy": 0.205218605697155,
|
|
"num_tokens": 11958340.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.23883843421936,
|
|
"epoch": 5.249886415265788,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002906090220905119,
|
|
"loss": 4.8051,
|
|
"mean_token_accuracy": 0.19460174292325974,
|
|
"num_tokens": 11968661.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.283464002609253,
|
|
"epoch": 5.254429804634257,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002902564253003677,
|
|
"loss": 4.7989,
|
|
"mean_token_accuracy": 0.19870449155569075,
|
|
"num_tokens": 11978282.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.28300371170044,
|
|
"epoch": 5.258973194002726,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002899037908665106,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.21340444535017014,
|
|
"num_tokens": 11987670.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.2956993103027346,
|
|
"epoch": 5.263516583371195,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002895511196590312,
|
|
"loss": 4.8787,
|
|
"mean_token_accuracy": 0.19468556791543962,
|
|
"num_tokens": 11999396.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.270407724380493,
|
|
"epoch": 5.268059972739664,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002891984125481104,
|
|
"loss": 4.8011,
|
|
"mean_token_accuracy": 0.20275041610002517,
|
|
"num_tokens": 12009894.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.310612154006958,
|
|
"epoch": 5.272603362108133,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00028884567040401814,
|
|
"loss": 4.8114,
|
|
"mean_token_accuracy": 0.2047180101275444,
|
|
"num_tokens": 12020132.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.274823093414307,
|
|
"epoch": 5.277146751476602,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002884928940971105,
|
|
"loss": 4.8016,
|
|
"mean_token_accuracy": 0.20106082558631896,
|
|
"num_tokens": 12030844.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.266018295288086,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000288140084497828,
|
|
"loss": 4.7989,
|
|
"mean_token_accuracy": 0.19389881044626237,
|
|
"num_tokens": 12040917.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.327526187896728,
|
|
"epoch": 5.286233530213539,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002877872424766932,
|
|
"loss": 4.8705,
|
|
"mean_token_accuracy": 0.20217728912830352,
|
|
"num_tokens": 12051443.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.317471551895141,
|
|
"epoch": 5.290776919582008,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002874343689043086,
|
|
"loss": 4.811,
|
|
"mean_token_accuracy": 0.21027474254369735,
|
|
"num_tokens": 12061383.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.267593812942505,
|
|
"epoch": 5.295320308950477,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000287081464651355,
|
|
"loss": 4.7582,
|
|
"mean_token_accuracy": 0.20562608689069747,
|
|
"num_tokens": 12070251.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.264637517929077,
|
|
"epoch": 5.299863698318946,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002867285305885882,
|
|
"loss": 4.8046,
|
|
"mean_token_accuracy": 0.20719066262245178,
|
|
"num_tokens": 12080828.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.406641817092895,
|
|
"epoch": 5.304407087687415,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028637556758683816,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.1990771159529686,
|
|
"num_tokens": 12091695.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.2882444858551025,
|
|
"epoch": 5.308950477055884,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002860225765170058,
|
|
"loss": 4.7741,
|
|
"mean_token_accuracy": 0.2108401358127594,
|
|
"num_tokens": 12103058.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.296488046646118,
|
|
"epoch": 5.313493866424353,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028566955825006174,
|
|
"loss": 4.8927,
|
|
"mean_token_accuracy": 0.1988106146454811,
|
|
"num_tokens": 12113026.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.238629150390625,
|
|
"epoch": 5.318037255792821,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00028531651365704325,
|
|
"loss": 4.7791,
|
|
"mean_token_accuracy": 0.20579945296049118,
|
|
"num_tokens": 12123055.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.3336341381073,
|
|
"epoch": 5.32258064516129,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00028496344360905287,
|
|
"loss": 4.8034,
|
|
"mean_token_accuracy": 0.2043280690908432,
|
|
"num_tokens": 12133538.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.274126720428467,
|
|
"epoch": 5.327124034529759,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00028461034897725587,
|
|
"loss": 4.7953,
|
|
"mean_token_accuracy": 0.20901448130607606,
|
|
"num_tokens": 12143840.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.250105857849121,
|
|
"epoch": 5.331667423898228,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028425723063287803,
|
|
"loss": 4.9036,
|
|
"mean_token_accuracy": 0.18952784836292266,
|
|
"num_tokens": 12154822.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.258183908462525,
|
|
"epoch": 5.336210813266697,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00028390408944720385,
|
|
"loss": 4.7764,
|
|
"mean_token_accuracy": 0.20804136693477632,
|
|
"num_tokens": 12163731.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.357541418075561,
|
|
"epoch": 5.340754202635166,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002835509262915741,
|
|
"loss": 4.8697,
|
|
"mean_token_accuracy": 0.19775439500808717,
|
|
"num_tokens": 12173574.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.302972602844238,
|
|
"epoch": 5.345297592003635,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00028319774203738375,
|
|
"loss": 4.7952,
|
|
"mean_token_accuracy": 0.20226332992315293,
|
|
"num_tokens": 12183456.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.234691333770752,
|
|
"epoch": 5.349840981372104,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002828445375560798,
|
|
"loss": 4.7726,
|
|
"mean_token_accuracy": 0.20865625739097596,
|
|
"num_tokens": 12192821.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.253103494644165,
|
|
"epoch": 5.354384370740572,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00028249131371915927,
|
|
"loss": 4.751,
|
|
"mean_token_accuracy": 0.2114056318998337,
|
|
"num_tokens": 12202904.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.28747444152832,
|
|
"epoch": 5.358927760109041,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002821380713981668,
|
|
"loss": 4.8317,
|
|
"mean_token_accuracy": 0.19875943660736084,
|
|
"num_tokens": 12213579.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.254388856887817,
|
|
"epoch": 5.36347114947751,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00028178481146469273,
|
|
"loss": 4.8723,
|
|
"mean_token_accuracy": 0.20323943346738815,
|
|
"num_tokens": 12223668.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.335781669616699,
|
|
"epoch": 5.368014538845979,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002814315347903709,
|
|
"loss": 4.9237,
|
|
"mean_token_accuracy": 0.20101699382066726,
|
|
"num_tokens": 12234869.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.274508619308472,
|
|
"epoch": 5.372557928214448,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002810782422468763,
|
|
"loss": 4.7668,
|
|
"mean_token_accuracy": 0.20194176435470582,
|
|
"num_tokens": 12246154.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.323515796661377,
|
|
"epoch": 5.377101317582917,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00028072493470592317,
|
|
"loss": 4.8407,
|
|
"mean_token_accuracy": 0.2049119621515274,
|
|
"num_tokens": 12256722.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.306603193283081,
|
|
"epoch": 5.381644706951386,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00028037161303926265,
|
|
"loss": 4.8122,
|
|
"mean_token_accuracy": 0.2084533080458641,
|
|
"num_tokens": 12266491.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.281565237045288,
|
|
"epoch": 5.386188096319855,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000280018278118681,
|
|
"loss": 4.8089,
|
|
"mean_token_accuracy": 0.20021868497133255,
|
|
"num_tokens": 12276946.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.266914701461792,
|
|
"epoch": 5.390731485688324,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000279664930815997,
|
|
"loss": 4.8106,
|
|
"mean_token_accuracy": 0.20558620393276214,
|
|
"num_tokens": 12286411.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.277288627624512,
|
|
"epoch": 5.395274875056792,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002793115720030599,
|
|
"loss": 4.8753,
|
|
"mean_token_accuracy": 0.20035961270332336,
|
|
"num_tokens": 12296215.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.3457457542419435,
|
|
"epoch": 5.399818264425261,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002789582025517474,
|
|
"loss": 4.8254,
|
|
"mean_token_accuracy": 0.20137866586446762,
|
|
"num_tokens": 12306246.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.295422554016113,
|
|
"epoch": 5.40436165379373,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0002786048233339637,
|
|
"loss": 4.8306,
|
|
"mean_token_accuracy": 0.20163779109716415,
|
|
"num_tokens": 12316824.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.252284526824951,
|
|
"epoch": 5.408905043162199,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00027825143522163675,
|
|
"loss": 4.823,
|
|
"mean_token_accuracy": 0.2114385962486267,
|
|
"num_tokens": 12326729.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.183179140090942,
|
|
"epoch": 5.413448432530668,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0002778980390867165,
|
|
"loss": 4.7222,
|
|
"mean_token_accuracy": 0.21433911323547364,
|
|
"num_tokens": 12335990.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.227345705032349,
|
|
"epoch": 5.417991821899137,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00027754463580117306,
|
|
"loss": 4.7677,
|
|
"mean_token_accuracy": 0.20771028846502304,
|
|
"num_tokens": 12345366.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.2586524963378904,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027719122623699353,
|
|
"loss": 4.88,
|
|
"mean_token_accuracy": 0.1998591959476471,
|
|
"num_tokens": 12355668.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.297056818008423,
|
|
"epoch": 5.427078600636074,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002768378112661813,
|
|
"loss": 4.8794,
|
|
"mean_token_accuracy": 0.19945432394742965,
|
|
"num_tokens": 12365885.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.307854700088501,
|
|
"epoch": 5.431621990004543,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002764843917607523,
|
|
"loss": 4.786,
|
|
"mean_token_accuracy": 0.20725948363542557,
|
|
"num_tokens": 12376795.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.251963520050049,
|
|
"epoch": 5.436165379373012,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002761309685927343,
|
|
"loss": 4.8069,
|
|
"mean_token_accuracy": 0.2123572811484337,
|
|
"num_tokens": 12386726.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.293879270553589,
|
|
"epoch": 5.440708768741481,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002757775426341636,
|
|
"loss": 4.8283,
|
|
"mean_token_accuracy": 0.19788987785577775,
|
|
"num_tokens": 12397264.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.318152713775635,
|
|
"epoch": 5.44525215810995,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00027542411475708385,
|
|
"loss": 4.8581,
|
|
"mean_token_accuracy": 0.2060815379023552,
|
|
"num_tokens": 12407744.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.247014141082763,
|
|
"epoch": 5.449795547478419,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027507068583354304,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.20145329535007478,
|
|
"num_tokens": 12418009.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.449795547478419,
|
|
"eval_entropy": 5.223560994170433,
|
|
"eval_loss": 5.764138221740723,
|
|
"eval_mean_token_accuracy": 0.16726842316430668,
|
|
"eval_num_tokens": 12418009.0,
|
|
"eval_runtime": 2.5869,
|
|
"eval_samples_per_second": 1327.849,
|
|
"eval_steps_per_second": 166.223,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.331464576721191,
|
|
"epoch": 5.454338936846888,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00027471725673559205,
|
|
"loss": 4.9019,
|
|
"mean_token_accuracy": 0.1953807845711708,
|
|
"num_tokens": 12427436.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.301974248886109,
|
|
"epoch": 5.458882326215357,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00027436382833528184,
|
|
"loss": 4.8637,
|
|
"mean_token_accuracy": 0.19851729571819304,
|
|
"num_tokens": 12438110.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.286835718154907,
|
|
"epoch": 5.463425715583826,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002740104015046621,
|
|
"loss": 4.8778,
|
|
"mean_token_accuracy": 0.20161221623420716,
|
|
"num_tokens": 12448428.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.374404954910278,
|
|
"epoch": 5.467969104952294,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027365697711577815,
|
|
"loss": 4.8535,
|
|
"mean_token_accuracy": 0.20768738240003587,
|
|
"num_tokens": 12459339.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.283126163482666,
|
|
"epoch": 5.472512494320763,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002733035560406697,
|
|
"loss": 4.8664,
|
|
"mean_token_accuracy": 0.19590149372816085,
|
|
"num_tokens": 12469723.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.18656964302063,
|
|
"epoch": 5.477055883689232,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027295013915136805,
|
|
"loss": 4.7529,
|
|
"mean_token_accuracy": 0.21057608872652053,
|
|
"num_tokens": 12480191.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.1912092685699465,
|
|
"epoch": 5.481599273057701,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00027259672731989436,
|
|
"loss": 4.7322,
|
|
"mean_token_accuracy": 0.21287806630134581,
|
|
"num_tokens": 12489881.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.22407169342041,
|
|
"epoch": 5.48614266242617,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000272243321418257,
|
|
"loss": 4.8162,
|
|
"mean_token_accuracy": 0.2035178080201149,
|
|
"num_tokens": 12499410.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.277679061889648,
|
|
"epoch": 5.490686051794639,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002718899223184501,
|
|
"loss": 4.8256,
|
|
"mean_token_accuracy": 0.2086278095841408,
|
|
"num_tokens": 12510079.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.262067699432373,
|
|
"epoch": 5.495229441163108,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00027153653089245067,
|
|
"loss": 4.8063,
|
|
"mean_token_accuracy": 0.20808187276124954,
|
|
"num_tokens": 12520093.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.263642644882202,
|
|
"epoch": 5.499772830531577,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000271183148012217,
|
|
"loss": 4.8597,
|
|
"mean_token_accuracy": 0.20227150470018387,
|
|
"num_tokens": 12530767.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.24763331413269,
|
|
"epoch": 5.504316219900045,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002708297745496861,
|
|
"loss": 4.7967,
|
|
"mean_token_accuracy": 0.20601402670145036,
|
|
"num_tokens": 12541218.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.226865148544311,
|
|
"epoch": 5.508859609268514,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00027047641137677213,
|
|
"loss": 4.7434,
|
|
"mean_token_accuracy": 0.20622434169054032,
|
|
"num_tokens": 12551112.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.279348993301392,
|
|
"epoch": 5.513402998636983,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002701230593653632,
|
|
"loss": 4.8518,
|
|
"mean_token_accuracy": 0.20152627676725388,
|
|
"num_tokens": 12561225.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.265758800506592,
|
|
"epoch": 5.517946388005452,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002697697193873207,
|
|
"loss": 4.7856,
|
|
"mean_token_accuracy": 0.2058170408010483,
|
|
"num_tokens": 12571508.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.20803050994873,
|
|
"epoch": 5.522489777373921,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026941639231447547,
|
|
"loss": 4.8672,
|
|
"mean_token_accuracy": 0.2022214949131012,
|
|
"num_tokens": 12582005.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.227892637252808,
|
|
"epoch": 5.52703316674239,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002690630790186273,
|
|
"loss": 4.8311,
|
|
"mean_token_accuracy": 0.21050778329372405,
|
|
"num_tokens": 12592243.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.265339946746826,
|
|
"epoch": 5.531576556110859,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026870978037154137,
|
|
"loss": 4.8316,
|
|
"mean_token_accuracy": 0.20178078413009642,
|
|
"num_tokens": 12603141.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.361912536621094,
|
|
"epoch": 5.536119945479328,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002683564972449472,
|
|
"loss": 4.9167,
|
|
"mean_token_accuracy": 0.19654146283864976,
|
|
"num_tokens": 12614811.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.336609601974487,
|
|
"epoch": 5.540663334847796,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002680032305105355,
|
|
"loss": 4.8924,
|
|
"mean_token_accuracy": 0.1975903570652008,
|
|
"num_tokens": 12624957.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.304868841171265,
|
|
"epoch": 5.545206724216265,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002676499810399572,
|
|
"loss": 4.8771,
|
|
"mean_token_accuracy": 0.19019340574741364,
|
|
"num_tokens": 12635898.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.250645112991333,
|
|
"epoch": 5.549750113584734,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00026729674970481985,
|
|
"loss": 4.8622,
|
|
"mean_token_accuracy": 0.20638727396726608,
|
|
"num_tokens": 12646176.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.246072769165039,
|
|
"epoch": 5.554293502953203,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00026694353737668705,
|
|
"loss": 4.7926,
|
|
"mean_token_accuracy": 0.20525217056274414,
|
|
"num_tokens": 12656226.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.238605117797851,
|
|
"epoch": 5.558836892321672,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026659034492707473,
|
|
"loss": 4.7951,
|
|
"mean_token_accuracy": 0.20731393992900848,
|
|
"num_tokens": 12666768.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.33120813369751,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002662371732274504,
|
|
"loss": 4.8322,
|
|
"mean_token_accuracy": 0.20734162181615828,
|
|
"num_tokens": 12676367.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.331324911117553,
|
|
"epoch": 5.56792367105861,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026588402314923,
|
|
"loss": 4.8608,
|
|
"mean_token_accuracy": 0.19886460155248642,
|
|
"num_tokens": 12687906.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.262137508392334,
|
|
"epoch": 5.572467060427079,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002655308955637763,
|
|
"loss": 4.8504,
|
|
"mean_token_accuracy": 0.20432738214731216,
|
|
"num_tokens": 12698236.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.262197637557984,
|
|
"epoch": 5.577010449795548,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002651777913423965,
|
|
"loss": 4.7704,
|
|
"mean_token_accuracy": 0.20792372077703475,
|
|
"num_tokens": 12707778.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.286218786239624,
|
|
"epoch": 5.581553839164016,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0002648247113563402,
|
|
"loss": 4.8088,
|
|
"mean_token_accuracy": 0.20187525749206542,
|
|
"num_tokens": 12717863.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.199344205856323,
|
|
"epoch": 5.586097228532485,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026447165647679703,
|
|
"loss": 4.8503,
|
|
"mean_token_accuracy": 0.2069487228989601,
|
|
"num_tokens": 12728061.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.24606671333313,
|
|
"epoch": 5.590640617900954,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000264118627574895,
|
|
"loss": 4.8571,
|
|
"mean_token_accuracy": 0.2056874543428421,
|
|
"num_tokens": 12738069.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.259573554992675,
|
|
"epoch": 5.595184007269423,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00026376562552169776,
|
|
"loss": 4.802,
|
|
"mean_token_accuracy": 0.20477283895015716,
|
|
"num_tokens": 12749024.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.261412620544434,
|
|
"epoch": 5.599727396637892,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00026341265118820284,
|
|
"loss": 4.818,
|
|
"mean_token_accuracy": 0.2015325054526329,
|
|
"num_tokens": 12757861.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.259226989746094,
|
|
"epoch": 5.604270786006361,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026305970544533903,
|
|
"loss": 4.8677,
|
|
"mean_token_accuracy": 0.20283048003911971,
|
|
"num_tokens": 12768108.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.301954221725464,
|
|
"epoch": 5.60881417537483,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026270678916396527,
|
|
"loss": 4.9367,
|
|
"mean_token_accuracy": 0.18905074745416642,
|
|
"num_tokens": 12778471.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.304304027557373,
|
|
"epoch": 5.613357564743298,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00026235390321486705,
|
|
"loss": 4.8796,
|
|
"mean_token_accuracy": 0.19948552697896957,
|
|
"num_tokens": 12789450.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.380365085601807,
|
|
"epoch": 5.617900954111767,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00026200104846875574,
|
|
"loss": 4.9361,
|
|
"mean_token_accuracy": 0.19338315725326538,
|
|
"num_tokens": 12800520.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.266437244415283,
|
|
"epoch": 5.622444343480236,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000261648225796265,
|
|
"loss": 4.7761,
|
|
"mean_token_accuracy": 0.2111216425895691,
|
|
"num_tokens": 12810587.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.213839340209961,
|
|
"epoch": 5.626987732848705,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002612954360679499,
|
|
"loss": 4.7951,
|
|
"mean_token_accuracy": 0.19939176142215728,
|
|
"num_tokens": 12820647.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.202492666244507,
|
|
"epoch": 5.631531122217174,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000260942680154284,
|
|
"loss": 4.8479,
|
|
"mean_token_accuracy": 0.20086337327957154,
|
|
"num_tokens": 12830863.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.339806652069091,
|
|
"epoch": 5.636074511585643,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002605899589256574,
|
|
"loss": 4.8892,
|
|
"mean_token_accuracy": 0.19383772015571593,
|
|
"num_tokens": 12842484.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.331115341186523,
|
|
"epoch": 5.640617900954112,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026023727325237463,
|
|
"loss": 4.8673,
|
|
"mean_token_accuracy": 0.19692640602588654,
|
|
"num_tokens": 12853009.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.2152101516723635,
|
|
"epoch": 5.645161290322581,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002598846240046527,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.21339505910873413,
|
|
"num_tokens": 12862726.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.258574676513672,
|
|
"epoch": 5.64970467969105,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002595320120526183,
|
|
"loss": 4.8167,
|
|
"mean_token_accuracy": 0.198949334025383,
|
|
"num_tokens": 12873389.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.222203540802002,
|
|
"epoch": 5.654248069059518,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00025917943826630666,
|
|
"loss": 4.7804,
|
|
"mean_token_accuracy": 0.2104671224951744,
|
|
"num_tokens": 12883221.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.264734554290771,
|
|
"epoch": 5.658791458427987,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002588269035156584,
|
|
"loss": 4.8772,
|
|
"mean_token_accuracy": 0.2018779918551445,
|
|
"num_tokens": 12893761.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.261203765869141,
|
|
"epoch": 5.663334847796456,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00025847440867051796,
|
|
"loss": 4.7828,
|
|
"mean_token_accuracy": 0.21167802065610886,
|
|
"num_tokens": 12903543.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.290173625946045,
|
|
"epoch": 5.667878237164925,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002581219546006314,
|
|
"loss": 4.8418,
|
|
"mean_token_accuracy": 0.20644614696502686,
|
|
"num_tokens": 12914330.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.310388231277466,
|
|
"epoch": 5.672421626533394,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002577695421756441,
|
|
"loss": 4.8897,
|
|
"mean_token_accuracy": 0.19068771600723267,
|
|
"num_tokens": 12925737.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.223596572875977,
|
|
"epoch": 5.676965015901863,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002574171722650986,
|
|
"loss": 4.773,
|
|
"mean_token_accuracy": 0.20758897364139556,
|
|
"num_tokens": 12935808.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.223459386825562,
|
|
"epoch": 5.681508405270332,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025706484573843273,
|
|
"loss": 4.7645,
|
|
"mean_token_accuracy": 0.21363472789525986,
|
|
"num_tokens": 12946358.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.264329719543457,
|
|
"epoch": 5.6860517946388,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000256712563464977,
|
|
"loss": 4.9008,
|
|
"mean_token_accuracy": 0.2039834663271904,
|
|
"num_tokens": 12956773.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.3024537563323975,
|
|
"epoch": 5.690595184007269,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002563603263139532,
|
|
"loss": 4.836,
|
|
"mean_token_accuracy": 0.19415371268987655,
|
|
"num_tokens": 12967381.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.295090818405152,
|
|
"epoch": 5.695138573375738,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002560081351544711,
|
|
"loss": 4.8621,
|
|
"mean_token_accuracy": 0.19801465272903443,
|
|
"num_tokens": 12977844.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.29942479133606,
|
|
"epoch": 5.699681962744207,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025565599085552776,
|
|
"loss": 4.8292,
|
|
"mean_token_accuracy": 0.20641437619924546,
|
|
"num_tokens": 12988180.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.212739086151123,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025530389428600394,
|
|
"loss": 4.8277,
|
|
"mean_token_accuracy": 0.19188137352466583,
|
|
"num_tokens": 12998651.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.199382543563843,
|
|
"epoch": 5.708768741481145,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002549518463146631,
|
|
"loss": 4.7738,
|
|
"mean_token_accuracy": 0.20636606067419053,
|
|
"num_tokens": 13009494.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.3285027027130125,
|
|
"epoch": 5.713312130849614,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002545998478101485,
|
|
"loss": 4.8156,
|
|
"mean_token_accuracy": 0.20050236731767654,
|
|
"num_tokens": 13019977.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.2700704574584964,
|
|
"epoch": 5.717855520218083,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025424789964098154,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.2045140042901039,
|
|
"num_tokens": 13030299.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.372756719589233,
|
|
"epoch": 5.722398909586552,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002538960026755593,
|
|
"loss": 4.9187,
|
|
"mean_token_accuracy": 0.2000368669629097,
|
|
"num_tokens": 13040339.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.296727418899536,
|
|
"epoch": 5.72694229895502,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00025354415778215255,
|
|
"loss": 4.8022,
|
|
"mean_token_accuracy": 0.20489950180053712,
|
|
"num_tokens": 13050187.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.272046327590942,
|
|
"epoch": 5.731485688323489,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00025319236582890364,
|
|
"loss": 4.8851,
|
|
"mean_token_accuracy": 0.20135241448879243,
|
|
"num_tokens": 13061067.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.274923801422119,
|
|
"epoch": 5.736029077691958,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025284062768382424,
|
|
"loss": 4.8531,
|
|
"mean_token_accuracy": 0.20571477562189103,
|
|
"num_tokens": 13071525.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.324580383300781,
|
|
"epoch": 5.740572467060427,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002524889442147932,
|
|
"loss": 4.857,
|
|
"mean_token_accuracy": 0.20698643177747728,
|
|
"num_tokens": 13081496.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.339061641693116,
|
|
"epoch": 5.745115856428896,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00025213731628955454,
|
|
"loss": 4.9041,
|
|
"mean_token_accuracy": 0.1964921087026596,
|
|
"num_tokens": 13091978.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.278643703460693,
|
|
"epoch": 5.749659245797365,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00025178574477571525,
|
|
"loss": 4.8178,
|
|
"mean_token_accuracy": 0.2015097036957741,
|
|
"num_tokens": 13102540.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.362884664535523,
|
|
"epoch": 5.754202635165834,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025143423054074305,
|
|
"loss": 4.8587,
|
|
"mean_token_accuracy": 0.20651526302099227,
|
|
"num_tokens": 13112794.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.313878011703491,
|
|
"epoch": 5.758746024534303,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002510827744519644,
|
|
"loss": 4.9408,
|
|
"mean_token_accuracy": 0.19590669274330139,
|
|
"num_tokens": 13123141.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.276621246337891,
|
|
"epoch": 5.763289413902772,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025073137737656226,
|
|
"loss": 4.8001,
|
|
"mean_token_accuracy": 0.21037306040525436,
|
|
"num_tokens": 13133410.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.259034490585327,
|
|
"epoch": 5.76783280327124,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002503800401815739,
|
|
"loss": 4.8499,
|
|
"mean_token_accuracy": 0.19911015182733535,
|
|
"num_tokens": 13143990.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.39494366645813,
|
|
"epoch": 5.772376192639709,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025002876373388913,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.1979763016104698,
|
|
"num_tokens": 13154664.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.323607873916626,
|
|
"epoch": 5.776919582008178,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024967754890024737,
|
|
"loss": 4.8784,
|
|
"mean_token_accuracy": 0.20080914497375488,
|
|
"num_tokens": 13165292.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.283810663223266,
|
|
"epoch": 5.781462971376647,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002493263965472366,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.19992492198944092,
|
|
"num_tokens": 13177111.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.304070329666137,
|
|
"epoch": 5.786006360745116,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0002489753075412901,
|
|
"loss": 4.8652,
|
|
"mean_token_accuracy": 0.1973637804389,
|
|
"num_tokens": 13186988.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.318563413619995,
|
|
"epoch": 5.790549750113585,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024862428274868533,
|
|
"loss": 4.8525,
|
|
"mean_token_accuracy": 0.21101430356502532,
|
|
"num_tokens": 13197338.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.301375675201416,
|
|
"epoch": 5.795093139482054,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002482733230355408,
|
|
"loss": 4.7744,
|
|
"mean_token_accuracy": 0.2072971373796463,
|
|
"num_tokens": 13208202.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.269202470779419,
|
|
"epoch": 5.799636528850522,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00024792242926781495,
|
|
"loss": 4.8207,
|
|
"mean_token_accuracy": 0.20681442618370055,
|
|
"num_tokens": 13218895.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.244685316085816,
|
|
"epoch": 5.804179918218991,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024757160231130313,
|
|
"loss": 4.8214,
|
|
"mean_token_accuracy": 0.20623739063739777,
|
|
"num_tokens": 13229060.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.270969438552856,
|
|
"epoch": 5.80872330758746,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024722084303163607,
|
|
"loss": 4.8475,
|
|
"mean_token_accuracy": 0.19909096658229827,
|
|
"num_tokens": 13239635.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.267392873764038,
|
|
"epoch": 5.813266696955929,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002468701522942774,
|
|
"loss": 4.8789,
|
|
"mean_token_accuracy": 0.19099488109350204,
|
|
"num_tokens": 13250783.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.2860846519470215,
|
|
"epoch": 5.817810086324398,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00024651953096452165,
|
|
"loss": 4.7651,
|
|
"mean_token_accuracy": 0.2121688559651375,
|
|
"num_tokens": 13260718.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.261371040344239,
|
|
"epoch": 5.822353475692867,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00024616897990749196,
|
|
"loss": 4.779,
|
|
"mean_token_accuracy": 0.2102072909474373,
|
|
"num_tokens": 13271606.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.2523821830749515,
|
|
"epoch": 5.826896865061336,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002458184999881385,
|
|
"loss": 4.799,
|
|
"mean_token_accuracy": 0.19869420528411866,
|
|
"num_tokens": 13281327.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.316252851486206,
|
|
"epoch": 5.831440254429805,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024546809207123534,
|
|
"loss": 4.8896,
|
|
"mean_token_accuracy": 0.19888270497322083,
|
|
"num_tokens": 13292871.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.243762969970703,
|
|
"epoch": 5.835983643798274,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002451177570213794,
|
|
"loss": 4.8102,
|
|
"mean_token_accuracy": 0.2023298442363739,
|
|
"num_tokens": 13303282.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.278167867660523,
|
|
"epoch": 5.840527033166742,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024476749570298746,
|
|
"loss": 4.8158,
|
|
"mean_token_accuracy": 0.2057469218969345,
|
|
"num_tokens": 13312987.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.276427459716797,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00024441730898029456,
|
|
"loss": 4.8827,
|
|
"mean_token_accuracy": 0.19989481568336487,
|
|
"num_tokens": 13324164.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.302016878128052,
|
|
"epoch": 5.84961381190368,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024406719771735165,
|
|
"loss": 4.8712,
|
|
"mean_token_accuracy": 0.19668232500553132,
|
|
"num_tokens": 13336790.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.3248261451721195,
|
|
"epoch": 5.854157201272149,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002437171627780234,
|
|
"loss": 4.8401,
|
|
"mean_token_accuracy": 0.2001558557152748,
|
|
"num_tokens": 13347107.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.258603239059449,
|
|
"epoch": 5.858700590640618,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024336720502598624,
|
|
"loss": 4.7747,
|
|
"mean_token_accuracy": 0.2161691591143608,
|
|
"num_tokens": 13356211.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.28527135848999,
|
|
"epoch": 5.863243980009087,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002430173253247262,
|
|
"loss": 4.772,
|
|
"mean_token_accuracy": 0.21711572259664536,
|
|
"num_tokens": 13366357.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.303808164596558,
|
|
"epoch": 5.867787369377556,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00024266752453753655,
|
|
"loss": 4.8658,
|
|
"mean_token_accuracy": 0.1959757536649704,
|
|
"num_tokens": 13375670.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.2679633617401125,
|
|
"epoch": 5.872330758746024,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024231780352751615,
|
|
"loss": 4.9081,
|
|
"mean_token_accuracy": 0.20292194783687592,
|
|
"num_tokens": 13387054.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.254691171646118,
|
|
"epoch": 5.876874148114493,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024196816315756665,
|
|
"loss": 4.7491,
|
|
"mean_token_accuracy": 0.21179336905479432,
|
|
"num_tokens": 13396314.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.305387401580811,
|
|
"epoch": 5.881417537482962,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002416186042903909,
|
|
"loss": 4.8591,
|
|
"mean_token_accuracy": 0.20255121886730193,
|
|
"num_tokens": 13405595.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.350368165969849,
|
|
"epoch": 5.885960926851431,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024126912778849074,
|
|
"loss": 4.9798,
|
|
"mean_token_accuracy": 0.1928052172064781,
|
|
"num_tokens": 13417008.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.3009075164794925,
|
|
"epoch": 5.8905043162199,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024091973451416472,
|
|
"loss": 4.8695,
|
|
"mean_token_accuracy": 0.2044842392206192,
|
|
"num_tokens": 13427650.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.280077266693115,
|
|
"epoch": 5.895047705588369,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024057042532950597,
|
|
"loss": 4.8696,
|
|
"mean_token_accuracy": 0.20722003430128097,
|
|
"num_tokens": 13437821.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.254697656631469,
|
|
"epoch": 5.899591094956838,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0002402212010964001,
|
|
"loss": 4.833,
|
|
"mean_token_accuracy": 0.19868321418762208,
|
|
"num_tokens": 13447685.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.297647953033447,
|
|
"epoch": 5.904134484325307,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00023987206267652334,
|
|
"loss": 4.8314,
|
|
"mean_token_accuracy": 0.20121184438467027,
|
|
"num_tokens": 13457929.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.904134484325307,
|
|
"eval_entropy": 5.244340374303419,
|
|
"eval_loss": 5.7301106452941895,
|
|
"eval_mean_token_accuracy": 0.16915378352237304,
|
|
"eval_num_tokens": 13457929.0,
|
|
"eval_runtime": 2.5888,
|
|
"eval_samples_per_second": 1326.887,
|
|
"eval_steps_per_second": 166.102,
|
|
"step": 6500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.949544505152e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|