3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.3625624716038165,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 4496087207808000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|