8079 lines
219 KiB
JSON
8079 lines
219 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.91434165442976,
|
|
"eval_steps": 1000,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691474437713623,
|
|
"epoch": 0.004894762604013705,
|
|
"grad_norm": 11.875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8265,
|
|
"mean_token_accuracy": 7.836990407668054e-05,
|
|
"num_tokens": 12593.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691561222076416,
|
|
"epoch": 0.00978952520802741,
|
|
"grad_norm": 11.75,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7489,
|
|
"mean_token_accuracy": 8.1300811143592e-05,
|
|
"num_tokens": 25410.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691450119018555,
|
|
"epoch": 0.014684287812041116,
|
|
"grad_norm": 9.6875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5176,
|
|
"mean_token_accuracy": 0.009127140190685168,
|
|
"num_tokens": 37264.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.688244438171386,
|
|
"epoch": 0.01957905041605482,
|
|
"grad_norm": 6.15625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2549,
|
|
"mean_token_accuracy": 0.030842235311865807,
|
|
"num_tokens": 49483.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.67002468109131,
|
|
"epoch": 0.024473813020068527,
|
|
"grad_norm": 4.15625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9838,
|
|
"mean_token_accuracy": 0.036930659972131255,
|
|
"num_tokens": 61479.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.649608993530274,
|
|
"epoch": 0.02936857562408223,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8397,
|
|
"mean_token_accuracy": 0.036012667790055276,
|
|
"num_tokens": 73344.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.640453147888184,
|
|
"epoch": 0.034263338228095935,
|
|
"grad_norm": 2.734375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7965,
|
|
"mean_token_accuracy": 0.04287212081253529,
|
|
"num_tokens": 85878.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.626565551757812,
|
|
"epoch": 0.03915810083210964,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7691,
|
|
"mean_token_accuracy": 0.04283439628779888,
|
|
"num_tokens": 97700.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.62145414352417,
|
|
"epoch": 0.04405286343612335,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6755,
|
|
"mean_token_accuracy": 0.04480269178748131,
|
|
"num_tokens": 109008.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.633209228515625,
|
|
"epoch": 0.048947626040137054,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6612,
|
|
"mean_token_accuracy": 0.04066384620964527,
|
|
"num_tokens": 120635.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.635868644714355,
|
|
"epoch": 0.05384238864415076,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5963,
|
|
"mean_token_accuracy": 0.04003451876342297,
|
|
"num_tokens": 133102.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.61568660736084,
|
|
"epoch": 0.05873715124816446,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4728,
|
|
"mean_token_accuracy": 0.050173386186361316,
|
|
"num_tokens": 145445.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.51130256652832,
|
|
"epoch": 0.06363191385217817,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3842,
|
|
"mean_token_accuracy": 0.05366942435503006,
|
|
"num_tokens": 157594.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.435814762115479,
|
|
"epoch": 0.06852667645619187,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2864,
|
|
"mean_token_accuracy": 0.06021526530385017,
|
|
"num_tokens": 170976.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.467036819458007,
|
|
"epoch": 0.07342143906020558,
|
|
"grad_norm": 2.875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2122,
|
|
"mean_token_accuracy": 0.06290490776300431,
|
|
"num_tokens": 183561.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.445936489105225,
|
|
"epoch": 0.07831620166421928,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0845,
|
|
"mean_token_accuracy": 0.07000090889632701,
|
|
"num_tokens": 195936.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.329971313476562,
|
|
"epoch": 0.08321096426823299,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9658,
|
|
"mean_token_accuracy": 0.07175729237496853,
|
|
"num_tokens": 208504.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.371177291870117,
|
|
"epoch": 0.0881057268722467,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9385,
|
|
"mean_token_accuracy": 0.06789587959647178,
|
|
"num_tokens": 219722.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.313321971893311,
|
|
"epoch": 0.0930004894762604,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.7436,
|
|
"mean_token_accuracy": 0.07605377212166786,
|
|
"num_tokens": 232307.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.276076507568359,
|
|
"epoch": 0.09789525208027411,
|
|
"grad_norm": 3.28125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.6775,
|
|
"mean_token_accuracy": 0.07455896399915218,
|
|
"num_tokens": 243698.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.196091747283935,
|
|
"epoch": 0.1027900146842878,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5575,
|
|
"mean_token_accuracy": 0.0785155214369297,
|
|
"num_tokens": 256418.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.124401664733886,
|
|
"epoch": 0.10768477728830152,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4104,
|
|
"mean_token_accuracy": 0.07805168032646179,
|
|
"num_tokens": 268503.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.06371488571167,
|
|
"epoch": 0.11257953989231523,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3101,
|
|
"mean_token_accuracy": 0.07913323007524013,
|
|
"num_tokens": 280386.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.980810928344727,
|
|
"epoch": 0.11747430249632893,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2019,
|
|
"mean_token_accuracy": 0.08205665312707425,
|
|
"num_tokens": 292121.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.89662504196167,
|
|
"epoch": 0.12236906510034264,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.0286,
|
|
"mean_token_accuracy": 0.08492446094751357,
|
|
"num_tokens": 303782.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.724525547027588,
|
|
"epoch": 0.12726382770435635,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9844,
|
|
"mean_token_accuracy": 0.08128817304968834,
|
|
"num_tokens": 315497.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.641802310943604,
|
|
"epoch": 0.13215859030837004,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9532,
|
|
"mean_token_accuracy": 0.07768743187189102,
|
|
"num_tokens": 327840.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.46103801727295,
|
|
"epoch": 0.13705335291238374,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.7654,
|
|
"mean_token_accuracy": 0.08798187449574471,
|
|
"num_tokens": 340192.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.228885173797607,
|
|
"epoch": 0.14194811551639747,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5669,
|
|
"mean_token_accuracy": 0.08950990438461304,
|
|
"num_tokens": 351561.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.089824771881103,
|
|
"epoch": 0.14684287812041116,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5159,
|
|
"mean_token_accuracy": 0.08800971060991288,
|
|
"num_tokens": 363668.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.857299709320069,
|
|
"epoch": 0.15173764072442486,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.4525,
|
|
"mean_token_accuracy": 0.09529466852545738,
|
|
"num_tokens": 375241.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.673880004882813,
|
|
"epoch": 0.15663240332843856,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.3172,
|
|
"mean_token_accuracy": 0.09652032777667045,
|
|
"num_tokens": 386488.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.434674930572509,
|
|
"epoch": 0.16152716593245228,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2006,
|
|
"mean_token_accuracy": 0.09962413907051086,
|
|
"num_tokens": 398203.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.35833911895752,
|
|
"epoch": 0.16642192853646598,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2842,
|
|
"mean_token_accuracy": 0.09556651711463929,
|
|
"num_tokens": 409381.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.216682815551758,
|
|
"epoch": 0.17131669114047968,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.2633,
|
|
"mean_token_accuracy": 0.09789874404668808,
|
|
"num_tokens": 421760.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.080347347259522,
|
|
"epoch": 0.1762114537444934,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1632,
|
|
"mean_token_accuracy": 0.10209440737962723,
|
|
"num_tokens": 432727.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.032786130905151,
|
|
"epoch": 0.1811062163485071,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.2471,
|
|
"mean_token_accuracy": 0.09438547641038894,
|
|
"num_tokens": 443799.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.938335800170899,
|
|
"epoch": 0.1860009789525208,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.1341,
|
|
"mean_token_accuracy": 0.10013610497117043,
|
|
"num_tokens": 455481.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.9153131484985355,
|
|
"epoch": 0.19089574155653452,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.0763,
|
|
"mean_token_accuracy": 0.09839447140693665,
|
|
"num_tokens": 467086.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.842766427993775,
|
|
"epoch": 0.19579050416054822,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.0763,
|
|
"mean_token_accuracy": 0.09429099857807159,
|
|
"num_tokens": 478227.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.8508209705352785,
|
|
"epoch": 0.20068526676456191,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.0819,
|
|
"mean_token_accuracy": 0.10272094458341599,
|
|
"num_tokens": 491036.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.750106430053711,
|
|
"epoch": 0.2055800293685756,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.0268,
|
|
"mean_token_accuracy": 0.10150075927376748,
|
|
"num_tokens": 502833.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.662298250198364,
|
|
"epoch": 0.21047479197258934,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.9994,
|
|
"mean_token_accuracy": 0.09913612380623818,
|
|
"num_tokens": 514078.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.76820936203003,
|
|
"epoch": 0.21536955457660303,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.0563,
|
|
"mean_token_accuracy": 0.09886371418833732,
|
|
"num_tokens": 525505.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.7181964874267575,
|
|
"epoch": 0.22026431718061673,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.0553,
|
|
"mean_token_accuracy": 0.10015972703695297,
|
|
"num_tokens": 536684.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.631637525558472,
|
|
"epoch": 0.22515907978463046,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.9678,
|
|
"mean_token_accuracy": 0.11216415911912918,
|
|
"num_tokens": 547664.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.560673618316651,
|
|
"epoch": 0.23005384238864415,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.9656,
|
|
"mean_token_accuracy": 0.10480529963970184,
|
|
"num_tokens": 559892.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.457711982727051,
|
|
"epoch": 0.23494860499265785,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.9001,
|
|
"mean_token_accuracy": 0.10354893431067466,
|
|
"num_tokens": 571012.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.583689403533936,
|
|
"epoch": 0.23984336759667158,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.9215,
|
|
"mean_token_accuracy": 0.10430897325277329,
|
|
"num_tokens": 583451.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.421637296676636,
|
|
"epoch": 0.24473813020068527,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9329,
|
|
"mean_token_accuracy": 0.10233990103006363,
|
|
"num_tokens": 595915.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.562687349319458,
|
|
"epoch": 0.24963289280469897,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.848,
|
|
"mean_token_accuracy": 0.10910093262791634,
|
|
"num_tokens": 607863.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.433994293212891,
|
|
"epoch": 0.2545276554087127,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.9084,
|
|
"mean_token_accuracy": 0.10604727566242218,
|
|
"num_tokens": 619306.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.461258888244629,
|
|
"epoch": 0.25942241801272636,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9468,
|
|
"mean_token_accuracy": 0.10922192186117172,
|
|
"num_tokens": 632553.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.41155481338501,
|
|
"epoch": 0.2643171806167401,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.7688,
|
|
"mean_token_accuracy": 0.11494352146983147,
|
|
"num_tokens": 643399.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.398804092407227,
|
|
"epoch": 0.2692119432207538,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.8403,
|
|
"mean_token_accuracy": 0.1158620223402977,
|
|
"num_tokens": 654148.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.385111284255982,
|
|
"epoch": 0.2741067058247675,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.8965,
|
|
"mean_token_accuracy": 0.11017908081412316,
|
|
"num_tokens": 666324.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.312957096099853,
|
|
"epoch": 0.2790014684287812,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8106,
|
|
"mean_token_accuracy": 0.11318352445960045,
|
|
"num_tokens": 677309.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.313508749008179,
|
|
"epoch": 0.28389623103279493,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8144,
|
|
"mean_token_accuracy": 0.10523941665887833,
|
|
"num_tokens": 691469.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.350642156600952,
|
|
"epoch": 0.2887909936368086,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8045,
|
|
"mean_token_accuracy": 0.12274143844842911,
|
|
"num_tokens": 702985.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.323756074905395,
|
|
"epoch": 0.2936857562408223,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.8789,
|
|
"mean_token_accuracy": 0.116054268181324,
|
|
"num_tokens": 715484.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.248594284057617,
|
|
"epoch": 0.29858051884483605,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7608,
|
|
"mean_token_accuracy": 0.11559304669499397,
|
|
"num_tokens": 725967.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.158112096786499,
|
|
"epoch": 0.3034752814488497,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.7019,
|
|
"mean_token_accuracy": 0.12221056818962098,
|
|
"num_tokens": 737878.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.245487976074219,
|
|
"epoch": 0.30837004405286345,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.7697,
|
|
"mean_token_accuracy": 0.1148179180920124,
|
|
"num_tokens": 750444.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.131648015975952,
|
|
"epoch": 0.3132648066568771,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.7355,
|
|
"mean_token_accuracy": 0.11609084904193878,
|
|
"num_tokens": 761905.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.189060926437378,
|
|
"epoch": 0.31815956926089084,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.6511,
|
|
"mean_token_accuracy": 0.11201457530260087,
|
|
"num_tokens": 773238.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.255410861968994,
|
|
"epoch": 0.32305433186490456,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.818,
|
|
"mean_token_accuracy": 0.11478052586317063,
|
|
"num_tokens": 785265.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.000382041931152,
|
|
"epoch": 0.32794909446891823,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.5323,
|
|
"mean_token_accuracy": 0.12431756779551506,
|
|
"num_tokens": 796214.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.132069635391235,
|
|
"epoch": 0.33284385707293196,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.7048,
|
|
"mean_token_accuracy": 0.12211181521415711,
|
|
"num_tokens": 807429.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.093725252151489,
|
|
"epoch": 0.3377386196769457,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7147,
|
|
"mean_token_accuracy": 0.11440666690468788,
|
|
"num_tokens": 819205.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.087794494628906,
|
|
"epoch": 0.34263338228095935,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.6921,
|
|
"mean_token_accuracy": 0.11648698076605797,
|
|
"num_tokens": 831417.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.148236560821533,
|
|
"epoch": 0.3475281448849731,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7295,
|
|
"mean_token_accuracy": 0.11438237279653549,
|
|
"num_tokens": 844792.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.161906480789185,
|
|
"epoch": 0.3524229074889868,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7167,
|
|
"mean_token_accuracy": 0.1205228291451931,
|
|
"num_tokens": 856643.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.08436393737793,
|
|
"epoch": 0.3573176700930005,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.6894,
|
|
"mean_token_accuracy": 0.1218317560851574,
|
|
"num_tokens": 867727.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.0502026081085205,
|
|
"epoch": 0.3622124326970142,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.6484,
|
|
"mean_token_accuracy": 0.12172565683722496,
|
|
"num_tokens": 879580.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.093457746505737,
|
|
"epoch": 0.3671071953010279,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6455,
|
|
"mean_token_accuracy": 0.11903778240084648,
|
|
"num_tokens": 890603.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.022216749191284,
|
|
"epoch": 0.3720019579050416,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.6953,
|
|
"mean_token_accuracy": 0.11619957983493805,
|
|
"num_tokens": 902566.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.156719017028808,
|
|
"epoch": 0.3768967205090553,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.7699,
|
|
"mean_token_accuracy": 0.11908866539597511,
|
|
"num_tokens": 915158.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.020287132263183,
|
|
"epoch": 0.38179148311306904,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.6569,
|
|
"mean_token_accuracy": 0.12016270384192466,
|
|
"num_tokens": 928299.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.987773513793945,
|
|
"epoch": 0.3866862457170827,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7094,
|
|
"mean_token_accuracy": 0.11790777072310447,
|
|
"num_tokens": 940470.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.102205657958985,
|
|
"epoch": 0.39158100832109644,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6851,
|
|
"mean_token_accuracy": 0.12131372764706612,
|
|
"num_tokens": 952661.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.911374616622925,
|
|
"epoch": 0.3964757709251101,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.6155,
|
|
"mean_token_accuracy": 0.11879359632730484,
|
|
"num_tokens": 965040.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.855518102645874,
|
|
"epoch": 0.40137053352912383,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5407,
|
|
"mean_token_accuracy": 0.12074877321720123,
|
|
"num_tokens": 977655.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.038438987731934,
|
|
"epoch": 0.40626529613313755,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.6291,
|
|
"mean_token_accuracy": 0.12360285148024559,
|
|
"num_tokens": 989422.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.930527019500732,
|
|
"epoch": 0.4111600587371512,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6136,
|
|
"mean_token_accuracy": 0.1211114652454853,
|
|
"num_tokens": 1002697.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.963910531997681,
|
|
"epoch": 0.41605482134116495,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.5715,
|
|
"mean_token_accuracy": 0.12396370768547058,
|
|
"num_tokens": 1014910.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.838316869735718,
|
|
"epoch": 0.4209495839451787,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.5537,
|
|
"mean_token_accuracy": 0.1339319072663784,
|
|
"num_tokens": 1026835.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.047098207473755,
|
|
"epoch": 0.42584434654919234,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6707,
|
|
"mean_token_accuracy": 0.11443433165550232,
|
|
"num_tokens": 1038792.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.900348567962647,
|
|
"epoch": 0.43073910915320607,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.6413,
|
|
"mean_token_accuracy": 0.11995069831609725,
|
|
"num_tokens": 1051335.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.856663560867309,
|
|
"epoch": 0.4356338717572198,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.5079,
|
|
"mean_token_accuracy": 0.130057705193758,
|
|
"num_tokens": 1063480.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.915047836303711,
|
|
"epoch": 0.44052863436123346,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.6642,
|
|
"mean_token_accuracy": 0.12786006629467012,
|
|
"num_tokens": 1074925.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.8480541706085205,
|
|
"epoch": 0.4454233969652472,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.396,
|
|
"mean_token_accuracy": 0.12921193838119507,
|
|
"num_tokens": 1086068.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.727921867370606,
|
|
"epoch": 0.4503181595692609,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.4969,
|
|
"mean_token_accuracy": 0.13089798390865326,
|
|
"num_tokens": 1097309.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.8872028350830075,
|
|
"epoch": 0.4552129221732746,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5581,
|
|
"mean_token_accuracy": 0.125684704631567,
|
|
"num_tokens": 1110253.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.802498722076416,
|
|
"epoch": 0.4601076847772883,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5386,
|
|
"mean_token_accuracy": 0.125102449208498,
|
|
"num_tokens": 1121297.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.830904102325439,
|
|
"epoch": 0.46500244738130203,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.4694,
|
|
"mean_token_accuracy": 0.13397082462906837,
|
|
"num_tokens": 1133378.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.764972782135009,
|
|
"epoch": 0.4698972099853157,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4908,
|
|
"mean_token_accuracy": 0.1247988536953926,
|
|
"num_tokens": 1146197.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.834934377670288,
|
|
"epoch": 0.4747919725893294,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.4895,
|
|
"mean_token_accuracy": 0.12815890833735466,
|
|
"num_tokens": 1158791.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.834185266494751,
|
|
"epoch": 0.47968673519334315,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4552,
|
|
"mean_token_accuracy": 0.13044697493314744,
|
|
"num_tokens": 1169977.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.6346056938171385,
|
|
"epoch": 0.4845814977973568,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.4207,
|
|
"mean_token_accuracy": 0.13247148767113687,
|
|
"num_tokens": 1182825.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.824855661392212,
|
|
"epoch": 0.48947626040137054,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5415,
|
|
"mean_token_accuracy": 0.12181857749819755,
|
|
"num_tokens": 1196101.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.662319135665894,
|
|
"epoch": 0.4943710230053842,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.3397,
|
|
"mean_token_accuracy": 0.1353870265185833,
|
|
"num_tokens": 1208916.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.704164838790893,
|
|
"epoch": 0.49926578560939794,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.3986,
|
|
"mean_token_accuracy": 0.13314119651913642,
|
|
"num_tokens": 1218972.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.779240608215332,
|
|
"epoch": 0.5041605482134116,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.13696231916546822,
|
|
"num_tokens": 1230692.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.710984706878662,
|
|
"epoch": 0.5090553108174254,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.3371,
|
|
"mean_token_accuracy": 0.14076263681054116,
|
|
"num_tokens": 1242104.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.6697978496551515,
|
|
"epoch": 0.5139500734214391,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.1359956182539463,
|
|
"num_tokens": 1254119.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.737369012832642,
|
|
"epoch": 0.5188448360254527,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4548,
|
|
"mean_token_accuracy": 0.1344248741865158,
|
|
"num_tokens": 1265272.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.730101156234741,
|
|
"epoch": 0.5237395986294665,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4222,
|
|
"mean_token_accuracy": 0.14020417779684066,
|
|
"num_tokens": 1277564.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.691190624237061,
|
|
"epoch": 0.5286343612334802,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.4491,
|
|
"mean_token_accuracy": 0.13473157808184624,
|
|
"num_tokens": 1289021.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.646734714508057,
|
|
"epoch": 0.5335291238374938,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3956,
|
|
"mean_token_accuracy": 0.13235612735152244,
|
|
"num_tokens": 1301101.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.719683647155762,
|
|
"epoch": 0.5384238864415076,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.5075,
|
|
"mean_token_accuracy": 0.1343328684568405,
|
|
"num_tokens": 1312982.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.649339962005615,
|
|
"epoch": 0.5433186490455213,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.298,
|
|
"mean_token_accuracy": 0.13852434232831,
|
|
"num_tokens": 1323587.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.660077142715454,
|
|
"epoch": 0.548213411649535,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4452,
|
|
"mean_token_accuracy": 0.13839047104120256,
|
|
"num_tokens": 1335989.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.759215688705444,
|
|
"epoch": 0.5531081742535487,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.47,
|
|
"mean_token_accuracy": 0.130010923743248,
|
|
"num_tokens": 1348065.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.673825311660766,
|
|
"epoch": 0.5580029368575624,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.3943,
|
|
"mean_token_accuracy": 0.13619856983423234,
|
|
"num_tokens": 1360005.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.64725513458252,
|
|
"epoch": 0.5628976994615761,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4037,
|
|
"mean_token_accuracy": 0.13103986009955407,
|
|
"num_tokens": 1372044.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.6476539134979244,
|
|
"epoch": 0.5677924620655899,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4819,
|
|
"mean_token_accuracy": 0.13560598343610764,
|
|
"num_tokens": 1384078.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.72802586555481,
|
|
"epoch": 0.5726872246696035,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.3781,
|
|
"mean_token_accuracy": 0.1315067872405052,
|
|
"num_tokens": 1396102.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.672314643859863,
|
|
"epoch": 0.5775819872736172,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.366,
|
|
"mean_token_accuracy": 0.13516665622591972,
|
|
"num_tokens": 1408009.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.647306442260742,
|
|
"epoch": 0.582476749877631,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3929,
|
|
"mean_token_accuracy": 0.13240321800112725,
|
|
"num_tokens": 1419883.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.443280220031738,
|
|
"epoch": 0.5873715124816447,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.1976,
|
|
"mean_token_accuracy": 0.1442122057080269,
|
|
"num_tokens": 1430901.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.578019142150879,
|
|
"epoch": 0.5922662750856583,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3251,
|
|
"mean_token_accuracy": 0.13777725845575334,
|
|
"num_tokens": 1443614.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.604838514328003,
|
|
"epoch": 0.5971610376896721,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.3186,
|
|
"mean_token_accuracy": 0.13596051931381226,
|
|
"num_tokens": 1454750.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.47489013671875,
|
|
"epoch": 0.6020558002936858,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.263,
|
|
"mean_token_accuracy": 0.13405164256691932,
|
|
"num_tokens": 1466424.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.579163694381714,
|
|
"epoch": 0.6069505628976994,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.3106,
|
|
"mean_token_accuracy": 0.1360399268567562,
|
|
"num_tokens": 1478762.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.509273624420166,
|
|
"epoch": 0.6118453255017132,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.3347,
|
|
"mean_token_accuracy": 0.13646816983819007,
|
|
"num_tokens": 1489683.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.514836597442627,
|
|
"epoch": 0.6167400881057269,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.2377,
|
|
"mean_token_accuracy": 0.13515424355864525,
|
|
"num_tokens": 1503307.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.6227192878723145,
|
|
"epoch": 0.6216348507097406,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3727,
|
|
"mean_token_accuracy": 0.13602764680981635,
|
|
"num_tokens": 1514982.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.397299718856812,
|
|
"epoch": 0.6265296133137542,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.2341,
|
|
"mean_token_accuracy": 0.14668046087026596,
|
|
"num_tokens": 1526760.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.560241603851319,
|
|
"epoch": 0.631424375917768,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.3118,
|
|
"mean_token_accuracy": 0.14421986192464828,
|
|
"num_tokens": 1537804.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.549531269073486,
|
|
"epoch": 0.6363191385217817,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3171,
|
|
"mean_token_accuracy": 0.1333357997238636,
|
|
"num_tokens": 1550423.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.398512887954712,
|
|
"epoch": 0.6412139011257953,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.218,
|
|
"mean_token_accuracy": 0.1467081144452095,
|
|
"num_tokens": 1563032.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.56570143699646,
|
|
"epoch": 0.6461086637298091,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3105,
|
|
"mean_token_accuracy": 0.13749707862734795,
|
|
"num_tokens": 1575600.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.476151990890503,
|
|
"epoch": 0.6510034263338228,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.2174,
|
|
"mean_token_accuracy": 0.1447758801281452,
|
|
"num_tokens": 1587604.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.376614093780518,
|
|
"epoch": 0.6558981889378365,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.24,
|
|
"mean_token_accuracy": 0.13755062445998192,
|
|
"num_tokens": 1599731.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.469626808166504,
|
|
"epoch": 0.6607929515418502,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.282,
|
|
"mean_token_accuracy": 0.1430655002593994,
|
|
"num_tokens": 1611944.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.544871091842651,
|
|
"epoch": 0.6656877141458639,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.1867,
|
|
"mean_token_accuracy": 0.1425486721098423,
|
|
"num_tokens": 1623177.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.3893975734710695,
|
|
"epoch": 0.6705824767498776,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2379,
|
|
"mean_token_accuracy": 0.14739967435598372,
|
|
"num_tokens": 1635075.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.408784580230713,
|
|
"epoch": 0.6754772393538914,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.1784,
|
|
"mean_token_accuracy": 0.14091105610132218,
|
|
"num_tokens": 1645126.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.488248634338379,
|
|
"epoch": 0.680372001957905,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.2174,
|
|
"mean_token_accuracy": 0.14224316030740738,
|
|
"num_tokens": 1656009.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.27689561843872,
|
|
"epoch": 0.6852667645619187,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.1975,
|
|
"mean_token_accuracy": 0.1452504150569439,
|
|
"num_tokens": 1667546.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.541668558120728,
|
|
"epoch": 0.6901615271659325,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.252,
|
|
"mean_token_accuracy": 0.13904105573892594,
|
|
"num_tokens": 1679176.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.330350160598755,
|
|
"epoch": 0.6950562897699462,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.1147,
|
|
"mean_token_accuracy": 0.1483779713511467,
|
|
"num_tokens": 1691240.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.385787868499756,
|
|
"epoch": 0.6999510523739598,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.1505,
|
|
"mean_token_accuracy": 0.1480160176753998,
|
|
"num_tokens": 1702185.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.384578990936279,
|
|
"epoch": 0.7048458149779736,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.1536,
|
|
"mean_token_accuracy": 0.14635321646928787,
|
|
"num_tokens": 1713702.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.361242246627808,
|
|
"epoch": 0.7097405775819873,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.2363,
|
|
"mean_token_accuracy": 0.14713326394557952,
|
|
"num_tokens": 1724943.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.3330967903137205,
|
|
"epoch": 0.714635340186001,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2454,
|
|
"mean_token_accuracy": 0.1357298105955124,
|
|
"num_tokens": 1737055.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.478425121307373,
|
|
"epoch": 0.7195301027900147,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2552,
|
|
"mean_token_accuracy": 0.13602484837174417,
|
|
"num_tokens": 1749772.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.379587030410766,
|
|
"epoch": 0.7244248653940284,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.2908,
|
|
"mean_token_accuracy": 0.13715352192521096,
|
|
"num_tokens": 1762533.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.489045238494873,
|
|
"epoch": 0.7293196279980421,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2885,
|
|
"mean_token_accuracy": 0.14545208886265754,
|
|
"num_tokens": 1774541.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.449005842208862,
|
|
"epoch": 0.7342143906020558,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2495,
|
|
"mean_token_accuracy": 0.14404454231262206,
|
|
"num_tokens": 1785627.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.345538234710693,
|
|
"epoch": 0.7391091532060695,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.074,
|
|
"mean_token_accuracy": 0.15339891761541366,
|
|
"num_tokens": 1796106.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.269445323944092,
|
|
"epoch": 0.7440039158100832,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.1115,
|
|
"mean_token_accuracy": 0.1419329449534416,
|
|
"num_tokens": 1808228.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.318842649459839,
|
|
"epoch": 0.748898678414097,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.1434,
|
|
"mean_token_accuracy": 0.1400843970477581,
|
|
"num_tokens": 1819415.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.33818244934082,
|
|
"epoch": 0.7537934410181106,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.1994,
|
|
"mean_token_accuracy": 0.14103697687387468,
|
|
"num_tokens": 1831043.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.344356966018677,
|
|
"epoch": 0.7586882036221243,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.1764,
|
|
"mean_token_accuracy": 0.13956040516495705,
|
|
"num_tokens": 1843062.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.430766439437866,
|
|
"epoch": 0.7635829662261381,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.2983,
|
|
"mean_token_accuracy": 0.1392583154141903,
|
|
"num_tokens": 1855181.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.23683648109436,
|
|
"epoch": 0.7684777288301518,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.0649,
|
|
"mean_token_accuracy": 0.147367200255394,
|
|
"num_tokens": 1866626.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.35269775390625,
|
|
"epoch": 0.7733724914341654,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.1448512777686119,
|
|
"num_tokens": 1878887.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.355936002731323,
|
|
"epoch": 0.7782672540381792,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.1022,
|
|
"mean_token_accuracy": 0.14598082453012468,
|
|
"num_tokens": 1890592.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.235448217391967,
|
|
"epoch": 0.7831620166421929,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.142,
|
|
"mean_token_accuracy": 0.14651560485363008,
|
|
"num_tokens": 1901875.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.273421049118042,
|
|
"epoch": 0.7880567792462065,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.1474,
|
|
"mean_token_accuracy": 0.13966845646500586,
|
|
"num_tokens": 1913913.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.30178804397583,
|
|
"epoch": 0.7929515418502202,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.0861,
|
|
"mean_token_accuracy": 0.1498040184378624,
|
|
"num_tokens": 1924667.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.291550588607788,
|
|
"epoch": 0.797846304454234,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.1604,
|
|
"mean_token_accuracy": 0.14530218094587327,
|
|
"num_tokens": 1935838.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.291401100158692,
|
|
"epoch": 0.8027410670582477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.1019,
|
|
"mean_token_accuracy": 0.15018822252750397,
|
|
"num_tokens": 1947608.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.300445222854615,
|
|
"epoch": 0.8076358296622613,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.1496323451399803,
|
|
"num_tokens": 1958426.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.2596992492675785,
|
|
"epoch": 0.8125305922662751,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.1174,
|
|
"mean_token_accuracy": 0.14941085800528525,
|
|
"num_tokens": 1971193.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.247623252868652,
|
|
"epoch": 0.8174253548702888,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.0576,
|
|
"mean_token_accuracy": 0.14635169357061387,
|
|
"num_tokens": 1982019.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.237645626068115,
|
|
"epoch": 0.8223201174743024,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1127,
|
|
"mean_token_accuracy": 0.15242150872945787,
|
|
"num_tokens": 1993991.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.244370412826538,
|
|
"epoch": 0.8272148800783162,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0823,
|
|
"mean_token_accuracy": 0.14876028150320053,
|
|
"num_tokens": 2006264.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.165139627456665,
|
|
"epoch": 0.8321096426823299,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1227,
|
|
"mean_token_accuracy": 0.15057290643453597,
|
|
"num_tokens": 2018308.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.253602123260498,
|
|
"epoch": 0.8370044052863436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.0878,
|
|
"mean_token_accuracy": 0.14493273943662643,
|
|
"num_tokens": 2030829.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.2643944263458256,
|
|
"epoch": 0.8418991678903573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.1606,
|
|
"mean_token_accuracy": 0.14317945316433905,
|
|
"num_tokens": 2041897.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.1602997303009035,
|
|
"epoch": 0.846793930494371,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.0424,
|
|
"mean_token_accuracy": 0.15380731970071793,
|
|
"num_tokens": 2054279.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.201865339279175,
|
|
"epoch": 0.8516886930983847,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1401,
|
|
"mean_token_accuracy": 0.1524582713842392,
|
|
"num_tokens": 2066150.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.292105579376221,
|
|
"epoch": 0.8565834557023985,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.1618,
|
|
"mean_token_accuracy": 0.1416109673678875,
|
|
"num_tokens": 2077895.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.222372436523438,
|
|
"epoch": 0.8614782183064121,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.0908,
|
|
"mean_token_accuracy": 0.1596553936600685,
|
|
"num_tokens": 2089205.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.336112594604492,
|
|
"epoch": 0.8663729809104258,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.187,
|
|
"mean_token_accuracy": 0.13830628544092177,
|
|
"num_tokens": 2099939.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.246317911148071,
|
|
"epoch": 0.8712677435144396,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.171,
|
|
"mean_token_accuracy": 0.1431927777826786,
|
|
"num_tokens": 2112148.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.198504447937012,
|
|
"epoch": 0.8761625061184533,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.0398,
|
|
"mean_token_accuracy": 0.1506725162267685,
|
|
"num_tokens": 2123375.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.207621812820435,
|
|
"epoch": 0.8810572687224669,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.1126,
|
|
"mean_token_accuracy": 0.14134914726018905,
|
|
"num_tokens": 2135884.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.248273086547852,
|
|
"epoch": 0.8859520313264807,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.1078,
|
|
"mean_token_accuracy": 0.14830219596624375,
|
|
"num_tokens": 2147912.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.193967342376709,
|
|
"epoch": 0.8908467939304944,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.0775,
|
|
"mean_token_accuracy": 0.1499609798192978,
|
|
"num_tokens": 2159683.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.1300129890441895,
|
|
"epoch": 0.895741556534508,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 5.9946,
|
|
"mean_token_accuracy": 0.1463859647512436,
|
|
"num_tokens": 2172528.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.162504768371582,
|
|
"epoch": 0.9006363191385218,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 5.9592,
|
|
"mean_token_accuracy": 0.15458274483680726,
|
|
"num_tokens": 2183947.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.154705381393432,
|
|
"epoch": 0.9055310817425355,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.0676,
|
|
"mean_token_accuracy": 0.1460274837911129,
|
|
"num_tokens": 2195167.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.221498584747314,
|
|
"epoch": 0.9104258443465492,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.0434,
|
|
"mean_token_accuracy": 0.1462668165564537,
|
|
"num_tokens": 2207830.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.170248174667359,
|
|
"epoch": 0.9153206069505629,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.0834,
|
|
"mean_token_accuracy": 0.15441582649946212,
|
|
"num_tokens": 2220717.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.178001356124878,
|
|
"epoch": 0.9202153695545766,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2167,
|
|
"mean_token_accuracy": 0.14381293207406998,
|
|
"num_tokens": 2232789.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.206983280181885,
|
|
"epoch": 0.9251101321585903,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0395,
|
|
"mean_token_accuracy": 0.14444569125771523,
|
|
"num_tokens": 2245155.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.121549892425537,
|
|
"epoch": 0.9300048947626041,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.012,
|
|
"mean_token_accuracy": 0.15058331415057183,
|
|
"num_tokens": 2256951.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.0408790588378904,
|
|
"epoch": 0.9348996573666177,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.0043,
|
|
"mean_token_accuracy": 0.15417952388525008,
|
|
"num_tokens": 2269087.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.1350022315979,
|
|
"epoch": 0.9397944199706314,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.0201,
|
|
"mean_token_accuracy": 0.1560334235429764,
|
|
"num_tokens": 2280563.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.190943622589112,
|
|
"epoch": 0.9446891825746452,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.2106,
|
|
"mean_token_accuracy": 0.14610329791903495,
|
|
"num_tokens": 2292454.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.0746870040893555,
|
|
"epoch": 0.9495839451786589,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 5.9563,
|
|
"mean_token_accuracy": 0.15315437838435172,
|
|
"num_tokens": 2305414.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.177819538116455,
|
|
"epoch": 0.9544787077826725,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.1509896457195282,
|
|
"num_tokens": 2316720.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.1112165451049805,
|
|
"epoch": 0.9593734703866863,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0576,
|
|
"mean_token_accuracy": 0.1500990182161331,
|
|
"num_tokens": 2330077.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.1100501537323,
|
|
"epoch": 0.9642682329907,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.0185,
|
|
"mean_token_accuracy": 0.14640207290649415,
|
|
"num_tokens": 2342943.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.120457935333252,
|
|
"epoch": 0.9691629955947136,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.0579,
|
|
"mean_token_accuracy": 0.15151562839746474,
|
|
"num_tokens": 2354770.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.153151655197144,
|
|
"epoch": 0.9740577581987273,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0659,
|
|
"mean_token_accuracy": 0.1489930845797062,
|
|
"num_tokens": 2367061.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.129442977905273,
|
|
"epoch": 0.9789525208027411,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 5.9541,
|
|
"mean_token_accuracy": 0.1546485498547554,
|
|
"num_tokens": 2379597.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9789525208027411,
|
|
"eval_entropy": 5.923030702807787,
|
|
"eval_loss": 6.252185821533203,
|
|
"eval_mean_token_accuracy": 0.14683223417214616,
|
|
"eval_num_tokens": 2379597.0,
|
|
"eval_runtime": 2.8719,
|
|
"eval_samples_per_second": 1200.257,
|
|
"eval_steps_per_second": 150.076,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.1020409107208256,
|
|
"epoch": 0.9838472834067548,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999997905632342,
|
|
"loss": 6.1611,
|
|
"mean_token_accuracy": 0.13938641250133516,
|
|
"num_tokens": 2391514.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.132669925689697,
|
|
"epoch": 0.9887420460107684,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999989397270414,
|
|
"loss": 6.0899,
|
|
"mean_token_accuracy": 0.14869284480810166,
|
|
"num_tokens": 2404156.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.21925802230835,
|
|
"epoch": 0.9936368086147822,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999974344040966,
|
|
"loss": 5.9988,
|
|
"mean_token_accuracy": 0.1483391046524048,
|
|
"num_tokens": 2415926.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.0971019744873045,
|
|
"epoch": 0.9985315712187959,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499995274598779,
|
|
"loss": 6.0348,
|
|
"mean_token_accuracy": 0.14810384511947633,
|
|
"num_tokens": 2428015.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 5.921160697937012,
|
|
"epoch": 1.0029368575624082,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999924603173707,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.15039441486199698,
|
|
"num_tokens": 2439258.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.056629610061646,
|
|
"epoch": 1.007831620166422,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999889915680581,
|
|
"loss": 5.8417,
|
|
"mean_token_accuracy": 0.15516380369663238,
|
|
"num_tokens": 2451155.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.037868022918701,
|
|
"epoch": 1.0127263827704356,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999848683609313,
|
|
"loss": 5.8081,
|
|
"mean_token_accuracy": 0.16687209755182267,
|
|
"num_tokens": 2462748.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 5.9815459728240965,
|
|
"epoch": 1.0176211453744493,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999800907079841,
|
|
"loss": 5.9096,
|
|
"mean_token_accuracy": 0.1518521785736084,
|
|
"num_tokens": 2476261.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.187918853759766,
|
|
"epoch": 1.0225159079784631,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999746586231139,
|
|
"loss": 5.8581,
|
|
"mean_token_accuracy": 0.15841317027807236,
|
|
"num_tokens": 2488913.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 5.942181348800659,
|
|
"epoch": 1.0274106705824768,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999685721221218,
|
|
"loss": 5.6992,
|
|
"mean_token_accuracy": 0.1668446958065033,
|
|
"num_tokens": 2499379.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.042421340942383,
|
|
"epoch": 1.0323054331864905,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999618312227125,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.15410418733954429,
|
|
"num_tokens": 2512713.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 5.975375127792359,
|
|
"epoch": 1.0372001957905042,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999544359444941,
|
|
"loss": 5.8255,
|
|
"mean_token_accuracy": 0.15900511890649796,
|
|
"num_tokens": 2524395.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 5.93801908493042,
|
|
"epoch": 1.0420949583945178,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999463863089784,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.1654394745826721,
|
|
"num_tokens": 2536182.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 5.901182365417481,
|
|
"epoch": 1.0469897209985315,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999376823395808,
|
|
"loss": 5.7942,
|
|
"mean_token_accuracy": 0.1684436872601509,
|
|
"num_tokens": 2548096.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.096840143203735,
|
|
"epoch": 1.0518844836025454,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999283240616193,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.15511096715927125,
|
|
"num_tokens": 2559770.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.056040048599243,
|
|
"epoch": 1.056779246206559,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999183115023159,
|
|
"loss": 5.8965,
|
|
"mean_token_accuracy": 0.1545817255973816,
|
|
"num_tokens": 2570826.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 5.975818204879761,
|
|
"epoch": 1.0616740088105727,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999076446907955,
|
|
"loss": 5.7988,
|
|
"mean_token_accuracy": 0.1615009590983391,
|
|
"num_tokens": 2582323.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 5.974529695510864,
|
|
"epoch": 1.0665687714145864,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998963236580864,
|
|
"loss": 5.8208,
|
|
"mean_token_accuracy": 0.15707745105028154,
|
|
"num_tokens": 2593855.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 5.976026439666748,
|
|
"epoch": 1.0714635340186,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998843484371195,
|
|
"loss": 5.8901,
|
|
"mean_token_accuracy": 0.15716765969991683,
|
|
"num_tokens": 2605396.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 5.932062435150146,
|
|
"epoch": 1.0763582966226137,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998717190627289,
|
|
"loss": 5.7899,
|
|
"mean_token_accuracy": 0.16322400867938996,
|
|
"num_tokens": 2617140.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 5.953844451904297,
|
|
"epoch": 1.0812530592266274,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998584355716515,
|
|
"loss": 5.8435,
|
|
"mean_token_accuracy": 0.15606771260499955,
|
|
"num_tokens": 2629901.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 5.946261262893676,
|
|
"epoch": 1.0861478218306413,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499844498002527,
|
|
"loss": 5.8081,
|
|
"mean_token_accuracy": 0.1612951472401619,
|
|
"num_tokens": 2643124.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 5.988902473449707,
|
|
"epoch": 1.091042584434655,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998299063958973,
|
|
"loss": 5.8323,
|
|
"mean_token_accuracy": 0.16525632739067078,
|
|
"num_tokens": 2656171.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 5.941033029556275,
|
|
"epoch": 1.0959373470386686,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004998146607942074,
|
|
"loss": 5.8125,
|
|
"mean_token_accuracy": 0.16049325913190843,
|
|
"num_tokens": 2668506.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 5.879610538482666,
|
|
"epoch": 1.1008321096426823,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004997987612418043,
|
|
"loss": 5.6705,
|
|
"mean_token_accuracy": 0.1725097879767418,
|
|
"num_tokens": 2681519.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 5.924249458312988,
|
|
"epoch": 1.105726872246696,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997822077849372,
|
|
"loss": 5.7649,
|
|
"mean_token_accuracy": 0.15810637921094894,
|
|
"num_tokens": 2694436.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 5.9003331661224365,
|
|
"epoch": 1.1106216348507096,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997650004717577,
|
|
"loss": 5.7727,
|
|
"mean_token_accuracy": 0.16366594582796096,
|
|
"num_tokens": 2707257.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 5.881779766082763,
|
|
"epoch": 1.1155163974547235,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997471393523191,
|
|
"loss": 5.7236,
|
|
"mean_token_accuracy": 0.16665565222501755,
|
|
"num_tokens": 2719934.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 5.840548467636109,
|
|
"epoch": 1.1204111600587372,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997286244785766,
|
|
"loss": 5.7338,
|
|
"mean_token_accuracy": 0.15967083722352982,
|
|
"num_tokens": 2732428.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 5.980451679229736,
|
|
"epoch": 1.1253059226627509,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004997094559043869,
|
|
"loss": 5.8326,
|
|
"mean_token_accuracy": 0.17002013325691223,
|
|
"num_tokens": 2743207.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 5.922123193740845,
|
|
"epoch": 1.1302006852667645,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004996896336855088,
|
|
"loss": 5.872,
|
|
"mean_token_accuracy": 0.16078334599733352,
|
|
"num_tokens": 2754451.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.043293762207031,
|
|
"epoch": 1.1350954478707782,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996691578796016,
|
|
"loss": 5.9171,
|
|
"mean_token_accuracy": 0.15479153990745545,
|
|
"num_tokens": 2765990.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 5.961138963699341,
|
|
"epoch": 1.1399902104747919,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996480285462265,
|
|
"loss": 5.8185,
|
|
"mean_token_accuracy": 0.16515018194913864,
|
|
"num_tokens": 2777376.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 5.919709777832031,
|
|
"epoch": 1.1448849730788058,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996262457468453,
|
|
"loss": 5.7851,
|
|
"mean_token_accuracy": 0.1699065536260605,
|
|
"num_tokens": 2788815.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.8117869853973385,
|
|
"epoch": 1.1497797356828194,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996038095448208,
|
|
"loss": 5.779,
|
|
"mean_token_accuracy": 0.16125653833150863,
|
|
"num_tokens": 2801138.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.98084454536438,
|
|
"epoch": 1.154674498286833,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995807200054164,
|
|
"loss": 5.792,
|
|
"mean_token_accuracy": 0.16341162621974945,
|
|
"num_tokens": 2813218.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.894933891296387,
|
|
"epoch": 1.1595692608908468,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000499556977195796,
|
|
"loss": 5.7365,
|
|
"mean_token_accuracy": 0.16507397592067719,
|
|
"num_tokens": 2825330.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.853562307357788,
|
|
"epoch": 1.1644640234948604,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995325811850237,
|
|
"loss": 5.8405,
|
|
"mean_token_accuracy": 0.15398580878973006,
|
|
"num_tokens": 2836606.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 5.856449317932129,
|
|
"epoch": 1.1693587860988741,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995075320440637,
|
|
"loss": 5.7043,
|
|
"mean_token_accuracy": 0.1697368621826172,
|
|
"num_tokens": 2847303.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.8915904521942135,
|
|
"epoch": 1.174253548702888,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004994818298457798,
|
|
"loss": 5.8025,
|
|
"mean_token_accuracy": 0.17078602463006973,
|
|
"num_tokens": 2858628.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 5.8079485416412355,
|
|
"epoch": 1.1791483113069017,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004994554746649361,
|
|
"loss": 5.7587,
|
|
"mean_token_accuracy": 0.16064557582139968,
|
|
"num_tokens": 2872074.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.937418842315674,
|
|
"epoch": 1.1840430739109153,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994284665781956,
|
|
"loss": 5.8741,
|
|
"mean_token_accuracy": 0.15438201725482942,
|
|
"num_tokens": 2884415.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.874923753738403,
|
|
"epoch": 1.188937836514929,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004994008056641203,
|
|
"loss": 5.7624,
|
|
"mean_token_accuracy": 0.16107115596532823,
|
|
"num_tokens": 2896867.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.8940307140350345,
|
|
"epoch": 1.1938325991189427,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993724920031716,
|
|
"loss": 5.7415,
|
|
"mean_token_accuracy": 0.16686908304691314,
|
|
"num_tokens": 2908378.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.846650695800781,
|
|
"epoch": 1.1987273617229564,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004993435256777098,
|
|
"loss": 5.7102,
|
|
"mean_token_accuracy": 0.17124667763710022,
|
|
"num_tokens": 2920124.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.995499849319458,
|
|
"epoch": 1.2036221243269702,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499313906771993,
|
|
"loss": 5.9352,
|
|
"mean_token_accuracy": 0.16055679321289062,
|
|
"num_tokens": 2931910.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.981334924697876,
|
|
"epoch": 1.208516886930984,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992836353721783,
|
|
"loss": 5.8286,
|
|
"mean_token_accuracy": 0.15923483520746232,
|
|
"num_tokens": 2944270.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.851262903213501,
|
|
"epoch": 1.2134116495349976,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004992527115663202,
|
|
"loss": 5.8178,
|
|
"mean_token_accuracy": 0.16358651891350745,
|
|
"num_tokens": 2955908.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.974487829208374,
|
|
"epoch": 1.2183064121390113,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992211354443714,
|
|
"loss": 5.8316,
|
|
"mean_token_accuracy": 0.15414459481835366,
|
|
"num_tokens": 2967432.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.881886148452759,
|
|
"epoch": 1.223201174743025,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499188907098182,
|
|
"loss": 5.8255,
|
|
"mean_token_accuracy": 0.1620217964053154,
|
|
"num_tokens": 2979092.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.85621337890625,
|
|
"epoch": 1.2280959373470386,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004991560266214989,
|
|
"loss": 5.7587,
|
|
"mean_token_accuracy": 0.16611319035291672,
|
|
"num_tokens": 2991993.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.932657051086426,
|
|
"epoch": 1.2329906999510523,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991224941099665,
|
|
"loss": 5.7767,
|
|
"mean_token_accuracy": 0.1694115564227104,
|
|
"num_tokens": 3004570.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.784896326065064,
|
|
"epoch": 1.2378854625550662,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004990883096611257,
|
|
"loss": 5.7954,
|
|
"mean_token_accuracy": 0.1682347059249878,
|
|
"num_tokens": 3016701.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.853497123718261,
|
|
"epoch": 1.2427802251590798,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004990534733744135,
|
|
"loss": 5.7733,
|
|
"mean_token_accuracy": 0.16693622022867202,
|
|
"num_tokens": 3028684.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.782205581665039,
|
|
"epoch": 1.2476749877630935,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004990179853511632,
|
|
"loss": 5.8238,
|
|
"mean_token_accuracy": 0.16249116659164428,
|
|
"num_tokens": 3040195.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.902102708816528,
|
|
"epoch": 1.2525697503671072,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989818456946042,
|
|
"loss": 5.6781,
|
|
"mean_token_accuracy": 0.1641119286417961,
|
|
"num_tokens": 3054195.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.842147397994995,
|
|
"epoch": 1.2574645129711208,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004989450545098609,
|
|
"loss": 5.7114,
|
|
"mean_token_accuracy": 0.16541883945465088,
|
|
"num_tokens": 3065621.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.671838665008545,
|
|
"epoch": 1.2623592755751347,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004989076119039532,
|
|
"loss": 5.6392,
|
|
"mean_token_accuracy": 0.1744843065738678,
|
|
"num_tokens": 3077973.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.917423486709595,
|
|
"epoch": 1.2672540381791484,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988695179857956,
|
|
"loss": 5.7565,
|
|
"mean_token_accuracy": 0.16196468621492385,
|
|
"num_tokens": 3091078.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.772269773483276,
|
|
"epoch": 1.272148800783162,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004988307728661976,
|
|
"loss": 5.6833,
|
|
"mean_token_accuracy": 0.17584044933319093,
|
|
"num_tokens": 3102823.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.8380790710449215,
|
|
"epoch": 1.2770435633871757,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004987913766578625,
|
|
"loss": 5.7022,
|
|
"mean_token_accuracy": 0.16909488290548325,
|
|
"num_tokens": 3113851.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.765717649459839,
|
|
"epoch": 1.2819383259911894,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987513294753879,
|
|
"loss": 5.7223,
|
|
"mean_token_accuracy": 0.168990258872509,
|
|
"num_tokens": 3125434.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.798812913894653,
|
|
"epoch": 1.286833088595203,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004987106314352647,
|
|
"loss": 5.6852,
|
|
"mean_token_accuracy": 0.1797244131565094,
|
|
"num_tokens": 3136908.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.832630443572998,
|
|
"epoch": 1.2917278511992167,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004986692826558771,
|
|
"loss": 5.6826,
|
|
"mean_token_accuracy": 0.1732470378279686,
|
|
"num_tokens": 3148378.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.767803382873535,
|
|
"epoch": 1.2966226138032306,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004986272832575023,
|
|
"loss": 5.6986,
|
|
"mean_token_accuracy": 0.16598847061395644,
|
|
"num_tokens": 3159612.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.879412412643433,
|
|
"epoch": 1.3015173764072443,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004985846333623098,
|
|
"loss": 5.8126,
|
|
"mean_token_accuracy": 0.16224471032619475,
|
|
"num_tokens": 3171855.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.839180994033813,
|
|
"epoch": 1.306412139011258,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004985413330943618,
|
|
"loss": 5.6453,
|
|
"mean_token_accuracy": 0.17059389799833297,
|
|
"num_tokens": 3184299.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.836867380142212,
|
|
"epoch": 1.3113069016152716,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984973825796119,
|
|
"loss": 5.7574,
|
|
"mean_token_accuracy": 0.16940534710884095,
|
|
"num_tokens": 3196015.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.72865138053894,
|
|
"epoch": 1.3162016642192853,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004984527819459051,
|
|
"loss": 5.5927,
|
|
"mean_token_accuracy": 0.17533381879329682,
|
|
"num_tokens": 3206525.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.784767293930054,
|
|
"epoch": 1.3210964268232992,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004984075313229779,
|
|
"loss": 5.7202,
|
|
"mean_token_accuracy": 0.16270063519477845,
|
|
"num_tokens": 3218117.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.805991840362549,
|
|
"epoch": 1.3259911894273126,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983616308424572,
|
|
"loss": 5.7103,
|
|
"mean_token_accuracy": 0.1695941358804703,
|
|
"num_tokens": 3229686.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.849427270889282,
|
|
"epoch": 1.3308859520313265,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983150806378601,
|
|
"loss": 5.7077,
|
|
"mean_token_accuracy": 0.17120572924613953,
|
|
"num_tokens": 3241779.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.816238069534302,
|
|
"epoch": 1.3357807146353402,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004982678808445943,
|
|
"loss": 5.7259,
|
|
"mean_token_accuracy": 0.16400495767593384,
|
|
"num_tokens": 3252033.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.842797422409058,
|
|
"epoch": 1.3406754772393539,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004982200315999563,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.16685152798891068,
|
|
"num_tokens": 3263987.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.683572149276733,
|
|
"epoch": 1.3455702398433675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004981715330431324,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.16725172400474547,
|
|
"num_tokens": 3275611.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.83439826965332,
|
|
"epoch": 1.3504650024473812,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981223853151969,
|
|
"loss": 5.7874,
|
|
"mean_token_accuracy": 0.16350995674729346,
|
|
"num_tokens": 3287620.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.767345333099366,
|
|
"epoch": 1.355359765051395,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004980725885591131,
|
|
"loss": 5.5957,
|
|
"mean_token_accuracy": 0.180044324696064,
|
|
"num_tokens": 3299651.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.723884010314942,
|
|
"epoch": 1.3602545276554088,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000498022142919732,
|
|
"loss": 5.7058,
|
|
"mean_token_accuracy": 0.16510243713855743,
|
|
"num_tokens": 3311619.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.72031536102295,
|
|
"epoch": 1.3651492902594224,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000497971048543792,
|
|
"loss": 5.6197,
|
|
"mean_token_accuracy": 0.17036764174699784,
|
|
"num_tokens": 3323658.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.776707029342651,
|
|
"epoch": 1.3700440528634361,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004979193055799187,
|
|
"loss": 5.67,
|
|
"mean_token_accuracy": 0.17030431926250458,
|
|
"num_tokens": 3335538.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.845314025878906,
|
|
"epoch": 1.3749388154674498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004978669141786242,
|
|
"loss": 5.8177,
|
|
"mean_token_accuracy": 0.1640164852142334,
|
|
"num_tokens": 3347424.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.7030088901519775,
|
|
"epoch": 1.3798335780714635,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004978138744923071,
|
|
"loss": 5.7139,
|
|
"mean_token_accuracy": 0.16582898795604706,
|
|
"num_tokens": 3359893.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.819153881072998,
|
|
"epoch": 1.3847283406754771,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004977601866752513,
|
|
"loss": 5.6884,
|
|
"mean_token_accuracy": 0.16975752115249634,
|
|
"num_tokens": 3371534.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.788328504562378,
|
|
"epoch": 1.389623103279491,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004977058508836266,
|
|
"loss": 5.6201,
|
|
"mean_token_accuracy": 0.17163756489753723,
|
|
"num_tokens": 3382912.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.59400749206543,
|
|
"epoch": 1.3945178658835047,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976508672754869,
|
|
"loss": 5.6404,
|
|
"mean_token_accuracy": 0.1698626771569252,
|
|
"num_tokens": 3394632.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.793394613265991,
|
|
"epoch": 1.3994126284875184,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004975952360107712,
|
|
"loss": 5.5841,
|
|
"mean_token_accuracy": 0.17525828778743743,
|
|
"num_tokens": 3405931.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.678069353103638,
|
|
"epoch": 1.404307391091532,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004975389572513023,
|
|
"loss": 5.6375,
|
|
"mean_token_accuracy": 0.1715028166770935,
|
|
"num_tokens": 3417944.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.730694150924682,
|
|
"epoch": 1.4092021536955457,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004974820311607858,
|
|
"loss": 5.7073,
|
|
"mean_token_accuracy": 0.16782306283712387,
|
|
"num_tokens": 3429949.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.879026794433594,
|
|
"epoch": 1.4140969162995596,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004974244579048114,
|
|
"loss": 5.7339,
|
|
"mean_token_accuracy": 0.16725998669862746,
|
|
"num_tokens": 3441951.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.70136284828186,
|
|
"epoch": 1.4189916789035733,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004973662376508503,
|
|
"loss": 5.7795,
|
|
"mean_token_accuracy": 0.1663120225071907,
|
|
"num_tokens": 3453807.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.853707551956177,
|
|
"epoch": 1.423886441507587,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004973073705682562,
|
|
"loss": 5.7488,
|
|
"mean_token_accuracy": 0.17351104617118834,
|
|
"num_tokens": 3465868.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.714973449707031,
|
|
"epoch": 1.4287812041116006,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004972478568282644,
|
|
"loss": 5.6466,
|
|
"mean_token_accuracy": 0.1747075527906418,
|
|
"num_tokens": 3477102.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.822602462768555,
|
|
"epoch": 1.4336759667156143,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004971876966039909,
|
|
"loss": 5.7085,
|
|
"mean_token_accuracy": 0.167814040184021,
|
|
"num_tokens": 3488845.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.673419380187989,
|
|
"epoch": 1.438570729319628,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004971268900704326,
|
|
"loss": 5.6544,
|
|
"mean_token_accuracy": 0.16964340806007386,
|
|
"num_tokens": 3500302.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.781837701797485,
|
|
"epoch": 1.4434654919236416,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004970654374044659,
|
|
"loss": 5.6533,
|
|
"mean_token_accuracy": 0.17830774933099747,
|
|
"num_tokens": 3512557.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.6502574443817135,
|
|
"epoch": 1.4483602545276555,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004970033387848472,
|
|
"loss": 5.6125,
|
|
"mean_token_accuracy": 0.1713570162653923,
|
|
"num_tokens": 3524415.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.85334095954895,
|
|
"epoch": 1.4532550171316692,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004969405943922116,
|
|
"loss": 5.7022,
|
|
"mean_token_accuracy": 0.17219314277172088,
|
|
"num_tokens": 3536321.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.714846754074097,
|
|
"epoch": 1.4581497797356828,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004968772044090728,
|
|
"loss": 5.6851,
|
|
"mean_token_accuracy": 0.1763856366276741,
|
|
"num_tokens": 3547606.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.656755828857422,
|
|
"epoch": 1.4630445423396965,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004968131690198224,
|
|
"loss": 5.5976,
|
|
"mean_token_accuracy": 0.1780870422720909,
|
|
"num_tokens": 3559935.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.717658948898316,
|
|
"epoch": 1.4679393049437102,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000496748488410729,
|
|
"loss": 5.6521,
|
|
"mean_token_accuracy": 0.1780029535293579,
|
|
"num_tokens": 3571022.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.818920993804932,
|
|
"epoch": 1.472834067547724,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000496683162769939,
|
|
"loss": 5.6748,
|
|
"mean_token_accuracy": 0.17060969322919844,
|
|
"num_tokens": 3582921.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.735677909851074,
|
|
"epoch": 1.4777288301517375,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004966171922874738,
|
|
"loss": 5.6074,
|
|
"mean_token_accuracy": 0.17647337913513184,
|
|
"num_tokens": 3593427.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.699115133285522,
|
|
"epoch": 1.4826235927557514,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004965505771552317,
|
|
"loss": 5.6062,
|
|
"mean_token_accuracy": 0.17297086268663406,
|
|
"num_tokens": 3605874.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.733571910858155,
|
|
"epoch": 1.487518355359765,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004964833175669854,
|
|
"loss": 5.6801,
|
|
"mean_token_accuracy": 0.1835954949259758,
|
|
"num_tokens": 3617475.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.592260789871216,
|
|
"epoch": 1.4924131179637787,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004964154137183828,
|
|
"loss": 5.6286,
|
|
"mean_token_accuracy": 0.17687239348888398,
|
|
"num_tokens": 3630634.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.758604955673218,
|
|
"epoch": 1.4973078805677924,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004963468658069456,
|
|
"loss": 5.6535,
|
|
"mean_token_accuracy": 0.16691459715366364,
|
|
"num_tokens": 3643384.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.680797386169433,
|
|
"epoch": 1.502202643171806,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004962776740320686,
|
|
"loss": 5.6159,
|
|
"mean_token_accuracy": 0.17650843262672425,
|
|
"num_tokens": 3654921.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.756531429290772,
|
|
"epoch": 1.50709740577582,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004962078385950203,
|
|
"loss": 5.6837,
|
|
"mean_token_accuracy": 0.17438367754220963,
|
|
"num_tokens": 3666959.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.671481561660767,
|
|
"epoch": 1.5119921683798334,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004961373596989407,
|
|
"loss": 5.6654,
|
|
"mean_token_accuracy": 0.17500680088996887,
|
|
"num_tokens": 3678227.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.7445825099945065,
|
|
"epoch": 1.5168869309838473,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004960662375488421,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.16987358331680297,
|
|
"num_tokens": 3690020.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.718591928482056,
|
|
"epoch": 1.521781693587861,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004959944723516077,
|
|
"loss": 5.6008,
|
|
"mean_token_accuracy": 0.1810409441590309,
|
|
"num_tokens": 3700619.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.781977605819702,
|
|
"epoch": 1.5266764561918746,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004959220643159912,
|
|
"loss": 5.6934,
|
|
"mean_token_accuracy": 0.1670381173491478,
|
|
"num_tokens": 3712532.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.714819812774659,
|
|
"epoch": 1.5315712187958885,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004958490136526163,
|
|
"loss": 5.6366,
|
|
"mean_token_accuracy": 0.17465935200452803,
|
|
"num_tokens": 3724454.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.647976398468018,
|
|
"epoch": 1.536465981399902,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000495775320573976,
|
|
"loss": 5.5818,
|
|
"mean_token_accuracy": 0.17660468220710754,
|
|
"num_tokens": 3735960.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.685937976837158,
|
|
"epoch": 1.5413607440039159,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004957009852944317,
|
|
"loss": 5.6185,
|
|
"mean_token_accuracy": 0.17408333122730255,
|
|
"num_tokens": 3748622.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.6516059875488285,
|
|
"epoch": 1.5462555066079295,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004956260080302134,
|
|
"loss": 5.5801,
|
|
"mean_token_accuracy": 0.17669832706451416,
|
|
"num_tokens": 3759597.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.741100454330445,
|
|
"epoch": 1.5511502692119432,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004955503889994183,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.16056070923805238,
|
|
"num_tokens": 3774056.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.6423286437988285,
|
|
"epoch": 1.5560450318159569,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00049547412842201,
|
|
"loss": 5.5775,
|
|
"mean_token_accuracy": 0.1796143189072609,
|
|
"num_tokens": 3786421.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.771421098709107,
|
|
"epoch": 1.5609397944199705,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004953972265198188,
|
|
"loss": 5.7024,
|
|
"mean_token_accuracy": 0.17257727086544036,
|
|
"num_tokens": 3799016.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.55297327041626,
|
|
"epoch": 1.5658345570239844,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004953196835165404,
|
|
"loss": 5.4507,
|
|
"mean_token_accuracy": 0.1835170716047287,
|
|
"num_tokens": 3810666.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.631848430633545,
|
|
"epoch": 1.5707293196279979,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000495241499637735,
|
|
"loss": 5.5193,
|
|
"mean_token_accuracy": 0.18205114156007768,
|
|
"num_tokens": 3820692.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.626948118209839,
|
|
"epoch": 1.5756240822320118,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004951626751108276,
|
|
"loss": 5.5369,
|
|
"mean_token_accuracy": 0.18134717494249344,
|
|
"num_tokens": 3831654.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.620755290985107,
|
|
"epoch": 1.5805188448360254,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.18002773821353912,
|
|
"num_tokens": 3842604.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.7996124744415285,
|
|
"epoch": 1.5854136074400391,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004950031050317222,
|
|
"loss": 5.6671,
|
|
"mean_token_accuracy": 0.17132557481527327,
|
|
"num_tokens": 3853991.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.596814584732056,
|
|
"epoch": 1.590308370044053,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004949223599436885,
|
|
"loss": 5.624,
|
|
"mean_token_accuracy": 0.17929036319255828,
|
|
"num_tokens": 3865173.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.714785194396972,
|
|
"epoch": 1.5952031326480665,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004948409751358803,
|
|
"loss": 5.6896,
|
|
"mean_token_accuracy": 0.16958404779434205,
|
|
"num_tokens": 3877777.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.675136232376099,
|
|
"epoch": 1.6000978952520803,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000494758950845033,
|
|
"loss": 5.7327,
|
|
"mean_token_accuracy": 0.17379774451255797,
|
|
"num_tokens": 3889944.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.749578189849854,
|
|
"epoch": 1.604992657856094,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004946762873097427,
|
|
"loss": 5.6411,
|
|
"mean_token_accuracy": 0.17379636764526368,
|
|
"num_tokens": 3901865.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.6178545475006105,
|
|
"epoch": 1.6098874204601077,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004945929847704646,
|
|
"loss": 5.5867,
|
|
"mean_token_accuracy": 0.1784010872244835,
|
|
"num_tokens": 3914694.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.674717617034912,
|
|
"epoch": 1.6147821830641214,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004945090434695127,
|
|
"loss": 5.5281,
|
|
"mean_token_accuracy": 0.18020234853029252,
|
|
"num_tokens": 3925446.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.6460856914520265,
|
|
"epoch": 1.619676945668135,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004944244636510592,
|
|
"loss": 5.6285,
|
|
"mean_token_accuracy": 0.17150038629770278,
|
|
"num_tokens": 3938216.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.621230363845825,
|
|
"epoch": 1.624571708272149,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004943392455611336,
|
|
"loss": 5.5389,
|
|
"mean_token_accuracy": 0.17987858653068542,
|
|
"num_tokens": 3950373.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.595808839797973,
|
|
"epoch": 1.6294664708761624,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000494253389447622,
|
|
"loss": 5.5604,
|
|
"mean_token_accuracy": 0.17657536119222642,
|
|
"num_tokens": 3961802.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.564799356460571,
|
|
"epoch": 1.6343612334801763,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004941668955602662,
|
|
"loss": 5.4352,
|
|
"mean_token_accuracy": 0.18312088549137115,
|
|
"num_tokens": 3973183.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.714397382736206,
|
|
"epoch": 1.63925599608419,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004940797641506638,
|
|
"loss": 5.6561,
|
|
"mean_token_accuracy": 0.1764741286635399,
|
|
"num_tokens": 3984556.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.6586535453796385,
|
|
"epoch": 1.6441507586882036,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004939919954722662,
|
|
"loss": 5.579,
|
|
"mean_token_accuracy": 0.17604130655527114,
|
|
"num_tokens": 3997195.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.608210277557373,
|
|
"epoch": 1.6490455212922175,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004939035897803787,
|
|
"loss": 5.5304,
|
|
"mean_token_accuracy": 0.17734461277723312,
|
|
"num_tokens": 4009921.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.6763317584991455,
|
|
"epoch": 1.653940283896231,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004938145473321597,
|
|
"loss": 5.7076,
|
|
"mean_token_accuracy": 0.1721693053841591,
|
|
"num_tokens": 4022109.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.685062742233276,
|
|
"epoch": 1.6588350465002448,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004937248683866198,
|
|
"loss": 5.663,
|
|
"mean_token_accuracy": 0.1764303296804428,
|
|
"num_tokens": 4035460.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.738089513778687,
|
|
"epoch": 1.6637298091042585,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004936345532046213,
|
|
"loss": 5.6089,
|
|
"mean_token_accuracy": 0.17306244969367982,
|
|
"num_tokens": 4047569.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.695468711853027,
|
|
"epoch": 1.6686245717082722,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004935436020488765,
|
|
"loss": 5.7061,
|
|
"mean_token_accuracy": 0.17170277535915374,
|
|
"num_tokens": 4061289.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.690058279037475,
|
|
"epoch": 1.6735193343122858,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004934520151839483,
|
|
"loss": 5.5552,
|
|
"mean_token_accuracy": 0.17911571711301805,
|
|
"num_tokens": 4073053.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.573972415924072,
|
|
"epoch": 1.6784140969162995,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004933597928762486,
|
|
"loss": 5.5514,
|
|
"mean_token_accuracy": 0.17718413323163987,
|
|
"num_tokens": 4084801.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.661252784729004,
|
|
"epoch": 1.6833088595203134,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004932669353940376,
|
|
"loss": 5.6629,
|
|
"mean_token_accuracy": 0.1697113633155823,
|
|
"num_tokens": 4097161.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.622225332260132,
|
|
"epoch": 1.6882036221243268,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004931734430074234,
|
|
"loss": 5.5034,
|
|
"mean_token_accuracy": 0.17641628682613372,
|
|
"num_tokens": 4108079.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.549112796783447,
|
|
"epoch": 1.6930983847283407,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004930793159883606,
|
|
"loss": 5.5236,
|
|
"mean_token_accuracy": 0.17827750146389007,
|
|
"num_tokens": 4119533.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.707393550872803,
|
|
"epoch": 1.6979931473323544,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004929845546106499,
|
|
"loss": 5.6569,
|
|
"mean_token_accuracy": 0.16999004334211348,
|
|
"num_tokens": 4132084.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.636761331558228,
|
|
"epoch": 1.702887909936368,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004928891591499373,
|
|
"loss": 5.7408,
|
|
"mean_token_accuracy": 0.16920279264450072,
|
|
"num_tokens": 4143113.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.820630693435669,
|
|
"epoch": 1.7077826725403817,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004927931298837135,
|
|
"loss": 5.6617,
|
|
"mean_token_accuracy": 0.1719266429543495,
|
|
"num_tokens": 4155774.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.592832279205322,
|
|
"epoch": 1.7126774351443954,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004926964670913126,
|
|
"loss": 5.5783,
|
|
"mean_token_accuracy": 0.176841701567173,
|
|
"num_tokens": 4167446.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.658400344848633,
|
|
"epoch": 1.7175721977484093,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004925991710539114,
|
|
"loss": 5.5782,
|
|
"mean_token_accuracy": 0.17812346816062927,
|
|
"num_tokens": 4179164.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.572030353546142,
|
|
"epoch": 1.7224669603524227,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004925012420545291,
|
|
"loss": 5.5317,
|
|
"mean_token_accuracy": 0.1798253834247589,
|
|
"num_tokens": 4189186.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.677495384216309,
|
|
"epoch": 1.7273617229564366,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004924026803780256,
|
|
"loss": 5.6403,
|
|
"mean_token_accuracy": 0.17903992384672165,
|
|
"num_tokens": 4200586.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.593063735961914,
|
|
"epoch": 1.7322564855604503,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004923034863111016,
|
|
"loss": 5.5394,
|
|
"mean_token_accuracy": 0.17907928675413132,
|
|
"num_tokens": 4211895.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.609043598175049,
|
|
"epoch": 1.737151248164464,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000492203660142297,
|
|
"loss": 5.5094,
|
|
"mean_token_accuracy": 0.18068955838680267,
|
|
"num_tokens": 4224143.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.654990005493164,
|
|
"epoch": 1.7420460107684779,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004921032021619905,
|
|
"loss": 5.6356,
|
|
"mean_token_accuracy": 0.1761782467365265,
|
|
"num_tokens": 4236435.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.6238361358642575,
|
|
"epoch": 1.7469407733724913,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492002112662399,
|
|
"loss": 5.623,
|
|
"mean_token_accuracy": 0.17351601719856263,
|
|
"num_tokens": 4247793.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.602977180480957,
|
|
"epoch": 1.7518355359765052,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004919003919375756,
|
|
"loss": 5.6,
|
|
"mean_token_accuracy": 0.17361711710691452,
|
|
"num_tokens": 4259739.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.6417151927948,
|
|
"epoch": 1.7567302985805189,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004917980402834104,
|
|
"loss": 5.5168,
|
|
"mean_token_accuracy": 0.18177273124456406,
|
|
"num_tokens": 4271720.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.639807510375976,
|
|
"epoch": 1.7616250611845325,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000491695057997628,
|
|
"loss": 5.589,
|
|
"mean_token_accuracy": 0.1812170147895813,
|
|
"num_tokens": 4285640.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.536652326583862,
|
|
"epoch": 1.7665198237885462,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004915914453797879,
|
|
"loss": 5.5274,
|
|
"mean_token_accuracy": 0.1794113725423813,
|
|
"num_tokens": 4296839.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.557571506500244,
|
|
"epoch": 1.7714145863925599,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004914872027312832,
|
|
"loss": 5.5082,
|
|
"mean_token_accuracy": 0.1859801158308983,
|
|
"num_tokens": 4309592.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.546999120712281,
|
|
"epoch": 1.7763093489965738,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004913823303553393,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.1869891345500946,
|
|
"num_tokens": 4320932.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.662409257888794,
|
|
"epoch": 1.7812041116005872,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004912768285570135,
|
|
"loss": 5.4952,
|
|
"mean_token_accuracy": 0.18336316645145417,
|
|
"num_tokens": 4332659.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.5181403160095215,
|
|
"epoch": 1.7860988742046011,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004911706976431942,
|
|
"loss": 5.526,
|
|
"mean_token_accuracy": 0.18275016844272612,
|
|
"num_tokens": 4343656.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.527674722671509,
|
|
"epoch": 1.7909936368086148,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004910639379225995,
|
|
"loss": 5.4257,
|
|
"mean_token_accuracy": 0.18914714008569716,
|
|
"num_tokens": 4354424.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.522308015823365,
|
|
"epoch": 1.7958883994126285,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004909565497057768,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.1798456832766533,
|
|
"num_tokens": 4365801.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.589231061935425,
|
|
"epoch": 1.8007831620166423,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004908485333051017,
|
|
"loss": 5.5345,
|
|
"mean_token_accuracy": 0.18014159202575683,
|
|
"num_tokens": 4377697.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.530683517456055,
|
|
"epoch": 1.8056779246206558,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000490739889034777,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.1751740276813507,
|
|
"num_tokens": 4389794.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.564450407028199,
|
|
"epoch": 1.8105726872246697,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004906306172108319,
|
|
"loss": 5.4342,
|
|
"mean_token_accuracy": 0.18271685540676116,
|
|
"num_tokens": 4403162.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.580508661270142,
|
|
"epoch": 1.8154674498286834,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000490520718151121,
|
|
"loss": 5.5508,
|
|
"mean_token_accuracy": 0.17712914496660231,
|
|
"num_tokens": 4416475.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.521164751052856,
|
|
"epoch": 1.820362212432697,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004904101921753234,
|
|
"loss": 5.5653,
|
|
"mean_token_accuracy": 0.17847047746181488,
|
|
"num_tokens": 4427229.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.590284538269043,
|
|
"epoch": 1.8252569750367107,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004902990396049421,
|
|
"loss": 5.5065,
|
|
"mean_token_accuracy": 0.18182663172483443,
|
|
"num_tokens": 4439609.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.556601619720459,
|
|
"epoch": 1.8301517376407244,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004901872607633024,
|
|
"loss": 5.538,
|
|
"mean_token_accuracy": 0.185947448015213,
|
|
"num_tokens": 4452870.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.595296812057495,
|
|
"epoch": 1.8350465002447383,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004900748559755517,
|
|
"loss": 5.5374,
|
|
"mean_token_accuracy": 0.17752565741539,
|
|
"num_tokens": 4464941.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.557303619384766,
|
|
"epoch": 1.8399412628487517,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004899618255686577,
|
|
"loss": 5.4161,
|
|
"mean_token_accuracy": 0.18646868616342543,
|
|
"num_tokens": 4475866.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.546024465560913,
|
|
"epoch": 1.8448360254527656,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004898481698714084,
|
|
"loss": 5.593,
|
|
"mean_token_accuracy": 0.179109650850296,
|
|
"num_tokens": 4487554.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.630650329589844,
|
|
"epoch": 1.8497307880567793,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004897338892144103,
|
|
"loss": 5.5629,
|
|
"mean_token_accuracy": 0.17396460026502608,
|
|
"num_tokens": 4499203.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.643454313278198,
|
|
"epoch": 1.854625550660793,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000489618983930088,
|
|
"loss": 5.6067,
|
|
"mean_token_accuracy": 0.17734497636556626,
|
|
"num_tokens": 4511567.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.544012451171875,
|
|
"epoch": 1.8595203132648068,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000489503454352683,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.17598173022270203,
|
|
"num_tokens": 4522898.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.588018608093262,
|
|
"epoch": 1.8644150758688203,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004893873008182531,
|
|
"loss": 5.5036,
|
|
"mean_token_accuracy": 0.18322323709726335,
|
|
"num_tokens": 4534289.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.626099157333374,
|
|
"epoch": 1.8693098384728342,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004892705236646702,
|
|
"loss": 5.5528,
|
|
"mean_token_accuracy": 0.17845290005207062,
|
|
"num_tokens": 4546073.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.5971643924713135,
|
|
"epoch": 1.8742046010768476,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004891531232316213,
|
|
"loss": 5.5567,
|
|
"mean_token_accuracy": 0.17655638009309768,
|
|
"num_tokens": 4557939.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.592231941223145,
|
|
"epoch": 1.8790993636808615,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004890350998606056,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.17622438818216324,
|
|
"num_tokens": 4570067.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.538221454620361,
|
|
"epoch": 1.8839941262848752,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004889164538949348,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.18144477158784866,
|
|
"num_tokens": 4581841.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.596781063079834,
|
|
"epoch": 1.8888888888888888,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004887971856797314,
|
|
"loss": 5.5239,
|
|
"mean_token_accuracy": 0.1778431847691536,
|
|
"num_tokens": 4594050.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.432804155349731,
|
|
"epoch": 1.8937836514929027,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000488677295561928,
|
|
"loss": 5.428,
|
|
"mean_token_accuracy": 0.18622596114873885,
|
|
"num_tokens": 4605456.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.587638425827026,
|
|
"epoch": 1.8986784140969162,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004885567838902663,
|
|
"loss": 5.5811,
|
|
"mean_token_accuracy": 0.17455296516418456,
|
|
"num_tokens": 4617110.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.579770183563232,
|
|
"epoch": 1.90357317670093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004884356510152958,
|
|
"loss": 5.5184,
|
|
"mean_token_accuracy": 0.1807307183742523,
|
|
"num_tokens": 4627172.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.510152912139892,
|
|
"epoch": 1.9084679393049437,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004883138972893732,
|
|
"loss": 5.4929,
|
|
"mean_token_accuracy": 0.18094931691884994,
|
|
"num_tokens": 4639710.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.590634489059449,
|
|
"epoch": 1.9133627019089574,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048819152306666096,
|
|
"loss": 5.4993,
|
|
"mean_token_accuracy": 0.18620602935552596,
|
|
"num_tokens": 4651526.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.541874361038208,
|
|
"epoch": 1.918257464512971,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048806852870312667,
|
|
"loss": 5.4362,
|
|
"mean_token_accuracy": 0.17709835022687911,
|
|
"num_tokens": 4662627.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.470085525512696,
|
|
"epoch": 1.9231522271169847,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004879449145565417,
|
|
"loss": 5.4582,
|
|
"mean_token_accuracy": 0.18288369923830033,
|
|
"num_tokens": 4674434.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.552283906936646,
|
|
"epoch": 1.9280469897209986,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048782068098648023,
|
|
"loss": 5.4158,
|
|
"mean_token_accuracy": 0.186358243227005,
|
|
"num_tokens": 4686028.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.587367248535156,
|
|
"epoch": 1.932941752325012,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004876958283543183,
|
|
"loss": 5.5802,
|
|
"mean_token_accuracy": 0.17691588252782822,
|
|
"num_tokens": 4698040.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.65945520401001,
|
|
"epoch": 1.937836514929026,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048757035702323275,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.1811575710773468,
|
|
"num_tokens": 4709274.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.476744174957275,
|
|
"epoch": 1.9427312775330396,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004874442673582001,
|
|
"loss": 5.5168,
|
|
"mean_token_accuracy": 0.18111482411623,
|
|
"num_tokens": 4722818.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.559481906890869,
|
|
"epoch": 1.9476260401370533,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004873175597259953,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.18057310134172438,
|
|
"num_tokens": 4733937.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.6205769062042235,
|
|
"epoch": 1.9525208027410672,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004871902344951913,
|
|
"loss": 5.5783,
|
|
"mean_token_accuracy": 0.17544569075107574,
|
|
"num_tokens": 4747473.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.592286682128906,
|
|
"epoch": 1.9574155653450807,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048706229203615716,
|
|
"loss": 5.6268,
|
|
"mean_token_accuracy": 0.17922073900699614,
|
|
"num_tokens": 4758675.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9574155653450807,
|
|
"eval_entropy": 5.535625642524241,
|
|
"eval_loss": 5.802816390991211,
|
|
"eval_mean_token_accuracy": 0.17496823080385368,
|
|
"eval_num_tokens": 4758675.0,
|
|
"eval_runtime": 2.9945,
|
|
"eval_samples_per_second": 1151.107,
|
|
"eval_steps_per_second": 143.93,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.626655435562133,
|
|
"epoch": 1.9623103279490945,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048693373272105743,
|
|
"loss": 5.504,
|
|
"mean_token_accuracy": 0.180687840282917,
|
|
"num_tokens": 4769947.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.595974683761597,
|
|
"epoch": 1.9672050905531082,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048680455692385124,
|
|
"loss": 5.5583,
|
|
"mean_token_accuracy": 0.18348373621702194,
|
|
"num_tokens": 4782749.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.579078435897827,
|
|
"epoch": 1.9720998531571219,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004866747650202907,
|
|
"loss": 5.5238,
|
|
"mean_token_accuracy": 0.18451536893844606,
|
|
"num_tokens": 4794944.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.6003326892852785,
|
|
"epoch": 1.9769946157611356,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004865443573879201,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.18887062072753907,
|
|
"num_tokens": 4806992.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.549536085128784,
|
|
"epoch": 1.9818893783651492,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004864133344060749,
|
|
"loss": 5.4539,
|
|
"mean_token_accuracy": 0.18131107091903687,
|
|
"num_tokens": 4818791.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.558773422241211,
|
|
"epoch": 1.9867841409691631,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048628169645588073,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.17917387038469315,
|
|
"num_tokens": 4830708.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.571465063095093,
|
|
"epoch": 1.9916789035731766,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004861494439202515,
|
|
"loss": 5.5524,
|
|
"mean_token_accuracy": 0.1800464168190956,
|
|
"num_tokens": 4843031.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.617353916168213,
|
|
"epoch": 1.9965736661771905,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004860165771838892,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.1778000995516777,
|
|
"num_tokens": 4855523.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.416062302059597,
|
|
"epoch": 2.0009789525208026,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004858830966332825,
|
|
"loss": 5.346,
|
|
"mean_token_accuracy": 0.18389874531163108,
|
|
"num_tokens": 4866979.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.488109588623047,
|
|
"epoch": 2.0058737151248165,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004857490026567055,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.19696198999881745,
|
|
"num_tokens": 4878496.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.498682832717895,
|
|
"epoch": 2.01076847772883,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004856142956442165,
|
|
"loss": 5.1833,
|
|
"mean_token_accuracy": 0.19493499547243118,
|
|
"num_tokens": 4890519.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.596047735214233,
|
|
"epoch": 2.015663240332844,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004854789759876572,
|
|
"loss": 5.2827,
|
|
"mean_token_accuracy": 0.19122138023376464,
|
|
"num_tokens": 4902797.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.531758785247803,
|
|
"epoch": 2.0205580029368577,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048534304408065135,
|
|
"loss": 5.1783,
|
|
"mean_token_accuracy": 0.19460100531578065,
|
|
"num_tokens": 4913243.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.424594974517822,
|
|
"epoch": 2.025452765540871,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048520650031860357,
|
|
"loss": 5.0638,
|
|
"mean_token_accuracy": 0.19740711152553558,
|
|
"num_tokens": 4923876.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.36385588645935,
|
|
"epoch": 2.030347528144885,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048506934509869845,
|
|
"loss": 5.0638,
|
|
"mean_token_accuracy": 0.2020087271928787,
|
|
"num_tokens": 4936004.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.394824838638305,
|
|
"epoch": 2.0352422907488985,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048493157881989894,
|
|
"loss": 5.098,
|
|
"mean_token_accuracy": 0.19910106658935547,
|
|
"num_tokens": 4947404.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.37768030166626,
|
|
"epoch": 2.0401370533529124,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004847932018829456,
|
|
"loss": 5.2171,
|
|
"mean_token_accuracy": 0.1941538408398628,
|
|
"num_tokens": 4960444.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.418075847625732,
|
|
"epoch": 2.0450318159569263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004846542146903555,
|
|
"loss": 5.1722,
|
|
"mean_token_accuracy": 0.19768768548965454,
|
|
"num_tokens": 4972356.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.449041891098022,
|
|
"epoch": 2.0499265785609397,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048451461764642046,
|
|
"loss": 5.1357,
|
|
"mean_token_accuracy": 0.19979992508888245,
|
|
"num_tokens": 4984785.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.291361522674561,
|
|
"epoch": 2.0548213411649536,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004843744111572067,
|
|
"loss": 5.0675,
|
|
"mean_token_accuracy": 0.1959679678082466,
|
|
"num_tokens": 4995909.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.432163047790527,
|
|
"epoch": 2.059716103768967,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004842335956305526,
|
|
"loss": 5.0949,
|
|
"mean_token_accuracy": 0.19766340106725694,
|
|
"num_tokens": 5006191.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.350412321090698,
|
|
"epoch": 2.064610866372981,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048409217147606895,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.2033711478114128,
|
|
"num_tokens": 5017399.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.316557788848877,
|
|
"epoch": 2.0695056289769944,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004839501391051363,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.19411368370056153,
|
|
"num_tokens": 5028402.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.446633148193359,
|
|
"epoch": 2.0744003915810083,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004838074989309045,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.20314160287380217,
|
|
"num_tokens": 5040813.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.414164018630982,
|
|
"epoch": 2.079295154185022,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048366425136829176,
|
|
"loss": 5.1433,
|
|
"mean_token_accuracy": 0.20220062136650085,
|
|
"num_tokens": 5053076.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.4324311256408695,
|
|
"epoch": 2.0841899167890356,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048352039683398286,
|
|
"loss": 5.2068,
|
|
"mean_token_accuracy": 0.19256707429885864,
|
|
"num_tokens": 5065397.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.374685382843017,
|
|
"epoch": 2.0890846793930495,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048337593574642796,
|
|
"loss": 5.0098,
|
|
"mean_token_accuracy": 0.20131478905677797,
|
|
"num_tokens": 5076545.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.4096685409545895,
|
|
"epoch": 2.093979441997063,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048323086852584187,
|
|
"loss": 5.2009,
|
|
"mean_token_accuracy": 0.19109852015972137,
|
|
"num_tokens": 5089487.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.413058519363403,
|
|
"epoch": 2.098874204601077,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004830851955942024,
|
|
"loss": 5.1923,
|
|
"mean_token_accuracy": 0.1943016231060028,
|
|
"num_tokens": 5102046.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.484387826919556,
|
|
"epoch": 2.1037689672050908,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004829389173752495,
|
|
"loss": 5.3399,
|
|
"mean_token_accuracy": 0.19494949281215668,
|
|
"num_tokens": 5114758.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.461303567886352,
|
|
"epoch": 2.108663729809104,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048279203429448366,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.19781216084957123,
|
|
"num_tokens": 5126130.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.358008193969726,
|
|
"epoch": 2.113558492413118,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004826445467791646,
|
|
"loss": 5.2108,
|
|
"mean_token_accuracy": 0.19490469843149186,
|
|
"num_tokens": 5137749.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.471472644805909,
|
|
"epoch": 2.1184532550171316,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048249645525831055,
|
|
"loss": 5.1601,
|
|
"mean_token_accuracy": 0.20036131292581558,
|
|
"num_tokens": 5149361.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.363424825668335,
|
|
"epoch": 2.1233480176211454,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004823477601626967,
|
|
"loss": 5.0896,
|
|
"mean_token_accuracy": 0.2008701056241989,
|
|
"num_tokens": 5161025.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.255881643295288,
|
|
"epoch": 2.128242780225159,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048219846192485365,
|
|
"loss": 5.0238,
|
|
"mean_token_accuracy": 0.20319584757089615,
|
|
"num_tokens": 5171507.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.427309846878051,
|
|
"epoch": 2.133137542829173,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004820485609790668,
|
|
"loss": 5.275,
|
|
"mean_token_accuracy": 0.19347049593925475,
|
|
"num_tokens": 5184026.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.4172220706939695,
|
|
"epoch": 2.1380323054331867,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048189805776137454,
|
|
"loss": 5.1795,
|
|
"mean_token_accuracy": 0.20001329630613326,
|
|
"num_tokens": 5195788.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.262393999099731,
|
|
"epoch": 2.1429270680372,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048174695270956707,
|
|
"loss": 5.0315,
|
|
"mean_token_accuracy": 0.20920827537775039,
|
|
"num_tokens": 5205440.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.293731689453125,
|
|
"epoch": 2.147821830641214,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004815952462631857,
|
|
"loss": 5.1003,
|
|
"mean_token_accuracy": 0.2001971885561943,
|
|
"num_tokens": 5216578.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.37099986076355,
|
|
"epoch": 2.1527165932452275,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048144293886352035,
|
|
"loss": 5.1344,
|
|
"mean_token_accuracy": 0.20005329102277755,
|
|
"num_tokens": 5228836.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.275912046432495,
|
|
"epoch": 2.1576113558492414,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004812900309536098,
|
|
"loss": 5.0887,
|
|
"mean_token_accuracy": 0.2067259818315506,
|
|
"num_tokens": 5240648.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.3612512111663815,
|
|
"epoch": 2.162506118453255,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048113652297823907,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.18981953859329223,
|
|
"num_tokens": 5252391.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.453209781646729,
|
|
"epoch": 2.1674008810572687,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048098241538393897,
|
|
"loss": 5.2036,
|
|
"mean_token_accuracy": 0.19229551404714584,
|
|
"num_tokens": 5264574.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.34055609703064,
|
|
"epoch": 2.1722956436612826,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048082770861898426,
|
|
"loss": 5.1746,
|
|
"mean_token_accuracy": 0.19398458898067475,
|
|
"num_tokens": 5277400.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.379359340667724,
|
|
"epoch": 2.177190406265296,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048067240313339295,
|
|
"loss": 5.1089,
|
|
"mean_token_accuracy": 0.19827474504709244,
|
|
"num_tokens": 5288951.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.3159558296203615,
|
|
"epoch": 2.18208516886931,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048051649937892423,
|
|
"loss": 5.0671,
|
|
"mean_token_accuracy": 0.2028965413570404,
|
|
"num_tokens": 5300902.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.325251483917237,
|
|
"epoch": 2.1869799314733234,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048035999780907785,
|
|
"loss": 5.1804,
|
|
"mean_token_accuracy": 0.1967838153243065,
|
|
"num_tokens": 5312498.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.33814377784729,
|
|
"epoch": 2.1918746940773373,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048020289887909246,
|
|
"loss": 5.0948,
|
|
"mean_token_accuracy": 0.19691168814897536,
|
|
"num_tokens": 5323591.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.343142366409301,
|
|
"epoch": 2.196769456681351,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004800452030459441,
|
|
"loss": 5.0878,
|
|
"mean_token_accuracy": 0.2060595616698265,
|
|
"num_tokens": 5334673.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.302957439422608,
|
|
"epoch": 2.2016642192853646,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004798869107683456,
|
|
"loss": 5.0975,
|
|
"mean_token_accuracy": 0.2082034319639206,
|
|
"num_tokens": 5347049.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.4850016117095945,
|
|
"epoch": 2.2065589818893785,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047972802250674426,
|
|
"loss": 5.2136,
|
|
"mean_token_accuracy": 0.1905737653374672,
|
|
"num_tokens": 5359153.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.326099538803101,
|
|
"epoch": 2.211453744493392,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047956853872332126,
|
|
"loss": 5.0928,
|
|
"mean_token_accuracy": 0.20605096518993377,
|
|
"num_tokens": 5369663.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.356387901306152,
|
|
"epoch": 2.216348507097406,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004794084598819899,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.1967753291130066,
|
|
"num_tokens": 5380962.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.288668155670166,
|
|
"epoch": 2.2212432697014193,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047924778644839466,
|
|
"loss": 5.1048,
|
|
"mean_token_accuracy": 0.2026848703622818,
|
|
"num_tokens": 5393319.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.390505695343018,
|
|
"epoch": 2.226138032305433,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047908651888990943,
|
|
"loss": 5.1442,
|
|
"mean_token_accuracy": 0.20073584467172623,
|
|
"num_tokens": 5405052.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.326527070999146,
|
|
"epoch": 2.231032794909447,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004789246576756362,
|
|
"loss": 5.0887,
|
|
"mean_token_accuracy": 0.2013887345790863,
|
|
"num_tokens": 5416416.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.414171504974365,
|
|
"epoch": 2.2359275575134605,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004787622032764041,
|
|
"loss": 5.3084,
|
|
"mean_token_accuracy": 0.18751744031906128,
|
|
"num_tokens": 5428676.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.395980930328369,
|
|
"epoch": 2.2408223201174744,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004785991561647675,
|
|
"loss": 5.1538,
|
|
"mean_token_accuracy": 0.19835059940814972,
|
|
"num_tokens": 5439885.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.331133651733398,
|
|
"epoch": 2.245717082721488,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047843551681500494,
|
|
"loss": 5.2068,
|
|
"mean_token_accuracy": 0.19880653768777848,
|
|
"num_tokens": 5451236.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.391649007797241,
|
|
"epoch": 2.2506118453255017,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004782712857031178,
|
|
"loss": 5.2206,
|
|
"mean_token_accuracy": 0.19829923510551453,
|
|
"num_tokens": 5464088.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.2941186904907225,
|
|
"epoch": 2.255506607929515,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004781064633068287,
|
|
"loss": 5.0972,
|
|
"mean_token_accuracy": 0.1968427300453186,
|
|
"num_tokens": 5475857.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.406572008132935,
|
|
"epoch": 2.260401370533529,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004779410501055802,
|
|
"loss": 5.2506,
|
|
"mean_token_accuracy": 0.19424178153276445,
|
|
"num_tokens": 5488670.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.382027053833008,
|
|
"epoch": 2.265296133137543,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00047777504658053343,
|
|
"loss": 5.1331,
|
|
"mean_token_accuracy": 0.19536806344985963,
|
|
"num_tokens": 5501746.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.342251920700074,
|
|
"epoch": 2.2701908957415564,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004776084532145668,
|
|
"loss": 5.2087,
|
|
"mean_token_accuracy": 0.19615187793970107,
|
|
"num_tokens": 5513504.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.28951063156128,
|
|
"epoch": 2.2750856583455703,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047744127049227425,
|
|
"loss": 5.1353,
|
|
"mean_token_accuracy": 0.1898852616548538,
|
|
"num_tokens": 5526144.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.376743125915527,
|
|
"epoch": 2.2799804209495838,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004772734988999644,
|
|
"loss": 5.1969,
|
|
"mean_token_accuracy": 0.19190049171447754,
|
|
"num_tokens": 5537545.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.363980293273926,
|
|
"epoch": 2.2848751835535976,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004771051389256582,
|
|
"loss": 5.1988,
|
|
"mean_token_accuracy": 0.19740142524242402,
|
|
"num_tokens": 5549447.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.373389053344726,
|
|
"epoch": 2.2897699461576115,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047693619105908887,
|
|
"loss": 5.1783,
|
|
"mean_token_accuracy": 0.19465951174497603,
|
|
"num_tokens": 5561091.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.344110679626465,
|
|
"epoch": 2.294664708761625,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004767666557916992,
|
|
"loss": 5.0986,
|
|
"mean_token_accuracy": 0.19979317635297775,
|
|
"num_tokens": 5573326.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.291253471374512,
|
|
"epoch": 2.299559471365639,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047659653361664087,
|
|
"loss": 5.1232,
|
|
"mean_token_accuracy": 0.19844507426023483,
|
|
"num_tokens": 5584947.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.330148124694825,
|
|
"epoch": 2.3044542339696523,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047642582502877256,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.20429294407367707,
|
|
"num_tokens": 5596944.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.328449010848999,
|
|
"epoch": 2.309348996573666,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000476254530524659,
|
|
"loss": 5.1795,
|
|
"mean_token_accuracy": 0.20087141394615174,
|
|
"num_tokens": 5609265.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.3663146018981935,
|
|
"epoch": 2.31424375917768,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004760826506025691,
|
|
"loss": 5.1304,
|
|
"mean_token_accuracy": 0.19760584831237793,
|
|
"num_tokens": 5620730.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.285108232498169,
|
|
"epoch": 2.3191385217816936,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004759101857624747,
|
|
"loss": 5.1766,
|
|
"mean_token_accuracy": 0.20640088468790055,
|
|
"num_tokens": 5633191.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.3933042049407955,
|
|
"epoch": 2.3240332843857074,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000475737136506049,
|
|
"loss": 5.2271,
|
|
"mean_token_accuracy": 0.1945857360959053,
|
|
"num_tokens": 5645134.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.295850324630737,
|
|
"epoch": 2.328928046989721,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047556350333666524,
|
|
"loss": 5.0834,
|
|
"mean_token_accuracy": 0.20512981414794923,
|
|
"num_tokens": 5658799.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.313294506072998,
|
|
"epoch": 2.333822809593735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004753892867593953,
|
|
"loss": 5.0859,
|
|
"mean_token_accuracy": 0.20842575281858444,
|
|
"num_tokens": 5670202.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.322448492050171,
|
|
"epoch": 2.3387175721977482,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047521448728100787,
|
|
"loss": 5.2334,
|
|
"mean_token_accuracy": 0.19291446208953858,
|
|
"num_tokens": 5682469.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.4270847797393795,
|
|
"epoch": 2.343612334801762,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047503910540996734,
|
|
"loss": 5.2345,
|
|
"mean_token_accuracy": 0.19537753313779832,
|
|
"num_tokens": 5692908.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.408088541030883,
|
|
"epoch": 2.348507097405776,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004748631416564322,
|
|
"loss": 5.1633,
|
|
"mean_token_accuracy": 0.20164895057678223,
|
|
"num_tokens": 5703813.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.196409845352173,
|
|
"epoch": 2.3534018600097895,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047468659653225343,
|
|
"loss": 4.9836,
|
|
"mean_token_accuracy": 0.21172523200511933,
|
|
"num_tokens": 5715274.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.29620771408081,
|
|
"epoch": 2.3582966226138034,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047450947055097326,
|
|
"loss": 5.1346,
|
|
"mean_token_accuracy": 0.1948849856853485,
|
|
"num_tokens": 5727809.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.345274257659912,
|
|
"epoch": 2.363191385217817,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004743317642278234,
|
|
"loss": 5.2698,
|
|
"mean_token_accuracy": 0.19395796507596968,
|
|
"num_tokens": 5740398.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.324441909790039,
|
|
"epoch": 2.3680861478218307,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000474153478079724,
|
|
"loss": 5.1448,
|
|
"mean_token_accuracy": 0.2060188740491867,
|
|
"num_tokens": 5752453.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.363028478622437,
|
|
"epoch": 2.372980910425844,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004739746126252813,
|
|
"loss": 5.2544,
|
|
"mean_token_accuracy": 0.19350223541259765,
|
|
"num_tokens": 5765062.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.430942630767822,
|
|
"epoch": 2.377875673029858,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004737951683847872,
|
|
"loss": 5.219,
|
|
"mean_token_accuracy": 0.19821250736713408,
|
|
"num_tokens": 5776765.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.291578006744385,
|
|
"epoch": 2.382770435633872,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004736151458802168,
|
|
"loss": 5.1678,
|
|
"mean_token_accuracy": 0.20081659257411957,
|
|
"num_tokens": 5789509.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.356848478317261,
|
|
"epoch": 2.3876651982378854,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004734345456352275,
|
|
"loss": 5.1043,
|
|
"mean_token_accuracy": 0.20367712676525115,
|
|
"num_tokens": 5801264.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.308737421035767,
|
|
"epoch": 2.3925599608418993,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004732533681751571,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.20206829607486726,
|
|
"num_tokens": 5813246.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.312044334411621,
|
|
"epoch": 2.3974547234459127,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004730716140270227,
|
|
"loss": 5.1469,
|
|
"mean_token_accuracy": 0.2028827801346779,
|
|
"num_tokens": 5825304.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.267190885543823,
|
|
"epoch": 2.4023494860499266,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004728892837195186,
|
|
"loss": 5.0377,
|
|
"mean_token_accuracy": 0.2059757873415947,
|
|
"num_tokens": 5837996.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.323007535934448,
|
|
"epoch": 2.4072442486539405,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004727063777830151,
|
|
"loss": 5.1748,
|
|
"mean_token_accuracy": 0.19384027421474456,
|
|
"num_tokens": 5850968.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.284120225906372,
|
|
"epoch": 2.412139011257954,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047252289674955706,
|
|
"loss": 5.1703,
|
|
"mean_token_accuracy": 0.1982756406068802,
|
|
"num_tokens": 5863233.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.306171417236328,
|
|
"epoch": 2.417033773861968,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047233884115286214,
|
|
"loss": 5.0615,
|
|
"mean_token_accuracy": 0.20489953309297562,
|
|
"num_tokens": 5873776.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.330395984649658,
|
|
"epoch": 2.4219285364659813,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004721542115283193,
|
|
"loss": 5.1998,
|
|
"mean_token_accuracy": 0.19702814221382142,
|
|
"num_tokens": 5885839.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.301304960250855,
|
|
"epoch": 2.426823299069995,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047196900841298725,
|
|
"loss": 5.1866,
|
|
"mean_token_accuracy": 0.1993207722902298,
|
|
"num_tokens": 5896973.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.327944040298462,
|
|
"epoch": 2.431718061674009,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004717832323455929,
|
|
"loss": 5.1753,
|
|
"mean_token_accuracy": 0.20232927352190017,
|
|
"num_tokens": 5908306.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.290309238433838,
|
|
"epoch": 2.4366128242780225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004715968838665298,
|
|
"loss": 5.1711,
|
|
"mean_token_accuracy": 0.19939935356378555,
|
|
"num_tokens": 5919719.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.33341794013977,
|
|
"epoch": 2.4415075868820364,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047140996351785654,
|
|
"loss": 5.1023,
|
|
"mean_token_accuracy": 0.20435553342103957,
|
|
"num_tokens": 5931406.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.368729209899902,
|
|
"epoch": 2.44640234948605,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047122247184329527,
|
|
"loss": 5.2185,
|
|
"mean_token_accuracy": 0.19418967366218567,
|
|
"num_tokens": 5943427.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.307469177246094,
|
|
"epoch": 2.4512971120900637,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004710344093882298,
|
|
"loss": 5.1507,
|
|
"mean_token_accuracy": 0.2013296440243721,
|
|
"num_tokens": 5954684.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.359532833099365,
|
|
"epoch": 2.456191874694077,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004708457766997046,
|
|
"loss": 5.2619,
|
|
"mean_token_accuracy": 0.18810701966285706,
|
|
"num_tokens": 5965135.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.419138956069946,
|
|
"epoch": 2.461086637298091,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047065657432642275,
|
|
"loss": 5.128,
|
|
"mean_token_accuracy": 0.20935332030057907,
|
|
"num_tokens": 5976419.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.281298351287842,
|
|
"epoch": 2.4659813999021045,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047046680281874416,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.19998786300420762,
|
|
"num_tokens": 5990204.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.401511573791504,
|
|
"epoch": 2.4708761625061184,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004702764627286845,
|
|
"loss": 5.2293,
|
|
"mean_token_accuracy": 0.19928793162107467,
|
|
"num_tokens": 6002137.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.285235929489136,
|
|
"epoch": 2.4757709251101323,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004700855546099136,
|
|
"loss": 5.1081,
|
|
"mean_token_accuracy": 0.20293399542570115,
|
|
"num_tokens": 6014254.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.33097767829895,
|
|
"epoch": 2.4806656877141458,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000469894079017753,
|
|
"loss": 5.1706,
|
|
"mean_token_accuracy": 0.1951598733663559,
|
|
"num_tokens": 6025999.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.242245769500732,
|
|
"epoch": 2.4855604503181596,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004697020365091755,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.2008503183722496,
|
|
"num_tokens": 6037909.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.338990831375122,
|
|
"epoch": 2.490455212922173,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004695094276428027,
|
|
"loss": 5.2176,
|
|
"mean_token_accuracy": 0.18787843734025955,
|
|
"num_tokens": 6050267.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.3056189060211185,
|
|
"epoch": 2.495349975526187,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046931625297890367,
|
|
"loss": 5.1037,
|
|
"mean_token_accuracy": 0.20342702567577362,
|
|
"num_tokens": 6062168.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.221565580368042,
|
|
"epoch": 2.500244738130201,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046912251307939336,
|
|
"loss": 5.063,
|
|
"mean_token_accuracy": 0.20233901143074035,
|
|
"num_tokens": 6075737.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.328143358230591,
|
|
"epoch": 2.5051395007342143,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004689282085078308,
|
|
"loss": 5.1588,
|
|
"mean_token_accuracy": 0.1983553498983383,
|
|
"num_tokens": 6089195.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.28541522026062,
|
|
"epoch": 2.510034263338228,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046873333982941763,
|
|
"loss": 5.1283,
|
|
"mean_token_accuracy": 0.19364329129457475,
|
|
"num_tokens": 6101277.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.3025566101074215,
|
|
"epoch": 2.5149290259422417,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004685379076109964,
|
|
"loss": 5.2084,
|
|
"mean_token_accuracy": 0.1892598196864128,
|
|
"num_tokens": 6113169.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.2775373458862305,
|
|
"epoch": 2.5198237885462555,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046834191242104887,
|
|
"loss": 5.0687,
|
|
"mean_token_accuracy": 0.20382051467895507,
|
|
"num_tokens": 6124006.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.243126773834229,
|
|
"epoch": 2.5247185511502694,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004681453548296944,
|
|
"loss": 5.0609,
|
|
"mean_token_accuracy": 0.20143638849258422,
|
|
"num_tokens": 6134991.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.288752269744873,
|
|
"epoch": 2.529613313754283,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004679482354086884,
|
|
"loss": 5.1545,
|
|
"mean_token_accuracy": 0.195332270860672,
|
|
"num_tokens": 6145771.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.329096794128418,
|
|
"epoch": 2.534508076358297,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046775055473142037,
|
|
"loss": 5.1409,
|
|
"mean_token_accuracy": 0.20373797565698623,
|
|
"num_tokens": 6157778.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.320698356628418,
|
|
"epoch": 2.5394028389623102,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046755231337291257,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.194973361492157,
|
|
"num_tokens": 6167977.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.258221435546875,
|
|
"epoch": 2.544297601566324,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000467353511909818,
|
|
"loss": 5.1519,
|
|
"mean_token_accuracy": 0.1974998250603676,
|
|
"num_tokens": 6179012.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.416901731491089,
|
|
"epoch": 2.549192364170338,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046715415092041923,
|
|
"loss": 5.2214,
|
|
"mean_token_accuracy": 0.20321713387966156,
|
|
"num_tokens": 6191199.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.396421337127686,
|
|
"epoch": 2.5540871267743515,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046695423098462607,
|
|
"loss": 5.2765,
|
|
"mean_token_accuracy": 0.19418736398220063,
|
|
"num_tokens": 6202990.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.373351144790649,
|
|
"epoch": 2.558981889378365,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004667537526839745,
|
|
"loss": 5.2863,
|
|
"mean_token_accuracy": 0.19325242787599564,
|
|
"num_tokens": 6215270.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.2843457698822025,
|
|
"epoch": 2.563876651982379,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004665527166016244,
|
|
"loss": 5.1041,
|
|
"mean_token_accuracy": 0.2006557270884514,
|
|
"num_tokens": 6227471.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.304417896270752,
|
|
"epoch": 2.5687714145863927,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004663511233223584,
|
|
"loss": 5.1405,
|
|
"mean_token_accuracy": 0.19751403629779815,
|
|
"num_tokens": 6239960.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.183145141601562,
|
|
"epoch": 2.573666177190406,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046614897343257993,
|
|
"loss": 5.0943,
|
|
"mean_token_accuracy": 0.19900689721107484,
|
|
"num_tokens": 6252986.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.334723329544067,
|
|
"epoch": 2.57856093979442,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004659462675203113,
|
|
"loss": 5.2365,
|
|
"mean_token_accuracy": 0.19438600391149521,
|
|
"num_tokens": 6265814.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.294690322875977,
|
|
"epoch": 2.5834557023984335,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004657430061751924,
|
|
"loss": 5.1526,
|
|
"mean_token_accuracy": 0.19875487089157104,
|
|
"num_tokens": 6276875.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.29907865524292,
|
|
"epoch": 2.5883504650024474,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046553918998847873,
|
|
"loss": 5.1866,
|
|
"mean_token_accuracy": 0.1913258284330368,
|
|
"num_tokens": 6288864.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.373501443862915,
|
|
"epoch": 2.5932452276064613,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004653348195530396,
|
|
"loss": 5.2424,
|
|
"mean_token_accuracy": 0.1908942475914955,
|
|
"num_tokens": 6302693.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.310032081604004,
|
|
"epoch": 2.5981399902104747,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046512989546335684,
|
|
"loss": 5.2109,
|
|
"mean_token_accuracy": 0.1932237148284912,
|
|
"num_tokens": 6315961.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.304839754104615,
|
|
"epoch": 2.6030347528144886,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046492441831552244,
|
|
"loss": 5.2087,
|
|
"mean_token_accuracy": 0.19596468061208724,
|
|
"num_tokens": 6328383.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.363293695449829,
|
|
"epoch": 2.607929515418502,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046471838870723744,
|
|
"loss": 5.136,
|
|
"mean_token_accuracy": 0.1972019836306572,
|
|
"num_tokens": 6341188.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.237611436843872,
|
|
"epoch": 2.612824278022516,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046451180723780966,
|
|
"loss": 5.0922,
|
|
"mean_token_accuracy": 0.20550482720136642,
|
|
"num_tokens": 6352364.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.364541006088257,
|
|
"epoch": 2.61771904062653,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046430467450815237,
|
|
"loss": 5.2227,
|
|
"mean_token_accuracy": 0.20022366791963578,
|
|
"num_tokens": 6364801.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.332408142089844,
|
|
"epoch": 2.6226138032305433,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00046409699112078234,
|
|
"loss": 5.1131,
|
|
"mean_token_accuracy": 0.20269023180007933,
|
|
"num_tokens": 6378350.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.227812242507935,
|
|
"epoch": 2.627508565834557,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000463888757679818,
|
|
"loss": 5.0975,
|
|
"mean_token_accuracy": 0.19431902468204498,
|
|
"num_tokens": 6390372.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.292713356018067,
|
|
"epoch": 2.6324033284385706,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000463679974790978,
|
|
"loss": 5.1655,
|
|
"mean_token_accuracy": 0.19646825790405273,
|
|
"num_tokens": 6402460.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.2531421184539795,
|
|
"epoch": 2.6372980910425845,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046347064306157904,
|
|
"loss": 5.0339,
|
|
"mean_token_accuracy": 0.20605625808238984,
|
|
"num_tokens": 6413086.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.273374080657959,
|
|
"epoch": 2.6421928536465984,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046326076310053443,
|
|
"loss": 5.2039,
|
|
"mean_token_accuracy": 0.20057330429553985,
|
|
"num_tokens": 6425334.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.329495000839233,
|
|
"epoch": 2.647087616250612,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004630503355183522,
|
|
"loss": 5.2021,
|
|
"mean_token_accuracy": 0.19525025635957718,
|
|
"num_tokens": 6438043.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.352862691879272,
|
|
"epoch": 2.6519823788546253,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004628393609271332,
|
|
"loss": 5.2473,
|
|
"mean_token_accuracy": 0.1973058670759201,
|
|
"num_tokens": 6449424.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.27777681350708,
|
|
"epoch": 2.656877141458639,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004626278399405698,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.20338010936975479,
|
|
"num_tokens": 6461149.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.37155704498291,
|
|
"epoch": 2.661771904062653,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004624157731739432,
|
|
"loss": 5.1824,
|
|
"mean_token_accuracy": 0.20031111240386962,
|
|
"num_tokens": 6472750.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.269566822052002,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046220316124412255,
|
|
"loss": 5.1534,
|
|
"mean_token_accuracy": 0.2030242934823036,
|
|
"num_tokens": 6485912.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.277011966705322,
|
|
"epoch": 2.6715614292706804,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004619900047695628,
|
|
"loss": 5.1074,
|
|
"mean_token_accuracy": 0.2049603432416916,
|
|
"num_tokens": 6496457.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.2734582901000975,
|
|
"epoch": 2.676456191874694,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004617763043703029,
|
|
"loss": 5.2227,
|
|
"mean_token_accuracy": 0.19709507524967193,
|
|
"num_tokens": 6509112.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.344844722747803,
|
|
"epoch": 2.6813509544787077,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004615620606679637,
|
|
"loss": 5.2023,
|
|
"mean_token_accuracy": 0.19780662953853606,
|
|
"num_tokens": 6520968.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.323253107070923,
|
|
"epoch": 2.6862457170827216,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004613472742857468,
|
|
"loss": 5.2199,
|
|
"mean_token_accuracy": 0.1946541726589203,
|
|
"num_tokens": 6533384.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.23217282295227,
|
|
"epoch": 2.691140479686735,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046113194584843217,
|
|
"loss": 5.08,
|
|
"mean_token_accuracy": 0.20902790129184723,
|
|
"num_tokens": 6544334.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.267884683609009,
|
|
"epoch": 2.696035242290749,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046091607598237655,
|
|
"loss": 5.1461,
|
|
"mean_token_accuracy": 0.20653503835201265,
|
|
"num_tokens": 6556260.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.357528781890869,
|
|
"epoch": 2.7009300048947624,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046069966531551174,
|
|
"loss": 5.1733,
|
|
"mean_token_accuracy": 0.1969723477959633,
|
|
"num_tokens": 6568148.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.397097110748291,
|
|
"epoch": 2.7058247674987763,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004604827144773425,
|
|
"loss": 5.2952,
|
|
"mean_token_accuracy": 0.19570557475090028,
|
|
"num_tokens": 6580929.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.2552718162536625,
|
|
"epoch": 2.71071953010279,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046026522409894476,
|
|
"loss": 5.2036,
|
|
"mean_token_accuracy": 0.19973676651716232,
|
|
"num_tokens": 6594412.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.272260332107544,
|
|
"epoch": 2.7156142927068037,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004600471948129643,
|
|
"loss": 5.1549,
|
|
"mean_token_accuracy": 0.20321643203496934,
|
|
"num_tokens": 6606735.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.34075813293457,
|
|
"epoch": 2.7205090553108175,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045982862725361414,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.19283341318368913,
|
|
"num_tokens": 6620203.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.281694984436035,
|
|
"epoch": 2.725403817914831,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045960952205667305,
|
|
"loss": 5.0273,
|
|
"mean_token_accuracy": 0.2054688349366188,
|
|
"num_tokens": 6632420.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.248138332366944,
|
|
"epoch": 2.730298580518845,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004593898798594839,
|
|
"loss": 5.0615,
|
|
"mean_token_accuracy": 0.20480649769306183,
|
|
"num_tokens": 6644090.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.18441481590271,
|
|
"epoch": 2.7351933431228588,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045916970130095155,
|
|
"loss": 5.1237,
|
|
"mean_token_accuracy": 0.20232576578855516,
|
|
"num_tokens": 6654970.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.235738801956177,
|
|
"epoch": 2.7400881057268722,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000458948987021541,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.20247414112091064,
|
|
"num_tokens": 6667428.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.293523788452148,
|
|
"epoch": 2.744982868330886,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004587277376632755,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.20096131265163422,
|
|
"num_tokens": 6679743.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.195648813247681,
|
|
"epoch": 2.7498776309348996,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004585059538697349,
|
|
"loss": 5.1231,
|
|
"mean_token_accuracy": 0.2077399328351021,
|
|
"num_tokens": 6691657.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.2407111644744875,
|
|
"epoch": 2.7547723935389135,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045828363628605355,
|
|
"loss": 5.1147,
|
|
"mean_token_accuracy": 0.20235977917909623,
|
|
"num_tokens": 6703862.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.314387226104737,
|
|
"epoch": 2.759667156142927,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045806078555891874,
|
|
"loss": 5.1477,
|
|
"mean_token_accuracy": 0.19421834498643875,
|
|
"num_tokens": 6714896.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.247271633148193,
|
|
"epoch": 2.764561918746941,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045783740233656825,
|
|
"loss": 5.1073,
|
|
"mean_token_accuracy": 0.20424874275922775,
|
|
"num_tokens": 6726432.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.23317141532898,
|
|
"epoch": 2.7694566813509542,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045761348726878906,
|
|
"loss": 5.0982,
|
|
"mean_token_accuracy": 0.20008147805929183,
|
|
"num_tokens": 6738262.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.314716863632202,
|
|
"epoch": 2.774351443954968,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004573890410069149,
|
|
"loss": 5.2182,
|
|
"mean_token_accuracy": 0.19755987972021102,
|
|
"num_tokens": 6750838.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.339499568939209,
|
|
"epoch": 2.779246206558982,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045716406420382507,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.20612797290086746,
|
|
"num_tokens": 6763284.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.200328302383423,
|
|
"epoch": 2.7841409691629955,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045693855751394185,
|
|
"loss": 5.1781,
|
|
"mean_token_accuracy": 0.19806490689516068,
|
|
"num_tokens": 6775122.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.246775674819946,
|
|
"epoch": 2.7890357317670094,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045671252159322907,
|
|
"loss": 4.9814,
|
|
"mean_token_accuracy": 0.21260356158018112,
|
|
"num_tokens": 6786126.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.225829267501831,
|
|
"epoch": 2.793930494371023,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045648595709918974,
|
|
"loss": 5.0955,
|
|
"mean_token_accuracy": 0.20787982046604156,
|
|
"num_tokens": 6797549.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.221702241897583,
|
|
"epoch": 2.7988252569750367,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004562588646908646,
|
|
"loss": 5.0589,
|
|
"mean_token_accuracy": 0.1999519497156143,
|
|
"num_tokens": 6809882.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.277024984359741,
|
|
"epoch": 2.8037200195790506,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045603124502883004,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.20202742516994476,
|
|
"num_tokens": 6822490.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.319039916992187,
|
|
"epoch": 2.808614782183064,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004558030987751961,
|
|
"loss": 5.1629,
|
|
"mean_token_accuracy": 0.19774821102619172,
|
|
"num_tokens": 6834857.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.291704940795898,
|
|
"epoch": 2.813509544787078,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004555744265936045,
|
|
"loss": 5.1726,
|
|
"mean_token_accuracy": 0.19582183808088302,
|
|
"num_tokens": 6847371.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.288819742202759,
|
|
"epoch": 2.8184043073910914,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004553452291492269,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.20653139501810075,
|
|
"num_tokens": 6858313.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.22146954536438,
|
|
"epoch": 2.8232990699951053,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045511550710876283,
|
|
"loss": 5.1053,
|
|
"mean_token_accuracy": 0.20446014255285264,
|
|
"num_tokens": 6869046.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.209698915481567,
|
|
"epoch": 2.828193832599119,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004548852611404378,
|
|
"loss": 4.9938,
|
|
"mean_token_accuracy": 0.21259300410747528,
|
|
"num_tokens": 6879484.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.295617961883545,
|
|
"epoch": 2.8330885952031326,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004546544919140014,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.2019681915640831,
|
|
"num_tokens": 6890941.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.305285978317261,
|
|
"epoch": 2.8379833578071465,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004544232001007253,
|
|
"loss": 5.1445,
|
|
"mean_token_accuracy": 0.19834502339363097,
|
|
"num_tokens": 6903446.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.2520472526550295,
|
|
"epoch": 2.84287812041116,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000454191386373401,
|
|
"loss": 5.139,
|
|
"mean_token_accuracy": 0.20072694271802902,
|
|
"num_tokens": 6916972.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.312392044067383,
|
|
"epoch": 2.847772883015174,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045395905140633875,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.19229765385389327,
|
|
"num_tokens": 6929888.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.216417598724365,
|
|
"epoch": 2.8526676456191877,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045372619587536427,
|
|
"loss": 5.0993,
|
|
"mean_token_accuracy": 0.20833698511123658,
|
|
"num_tokens": 6941739.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.288608837127685,
|
|
"epoch": 2.857562408223201,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004534928204578182,
|
|
"loss": 5.1386,
|
|
"mean_token_accuracy": 0.20470602214336395,
|
|
"num_tokens": 6953842.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.276420736312867,
|
|
"epoch": 2.8624571708272146,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004532589258325531,
|
|
"loss": 5.1236,
|
|
"mean_token_accuracy": 0.20376610457897187,
|
|
"num_tokens": 6965217.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.278570938110351,
|
|
"epoch": 2.8673519334312285,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004530245126799317,
|
|
"loss": 5.1017,
|
|
"mean_token_accuracy": 0.20043444782495498,
|
|
"num_tokens": 6976996.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.156794118881225,
|
|
"epoch": 2.8722466960352424,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004527895816818255,
|
|
"loss": 5.077,
|
|
"mean_token_accuracy": 0.2072145789861679,
|
|
"num_tokens": 6987481.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.305797386169433,
|
|
"epoch": 2.877141458639256,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004525541335216119,
|
|
"loss": 5.1421,
|
|
"mean_token_accuracy": 0.2094895660877228,
|
|
"num_tokens": 6998297.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.271157121658325,
|
|
"epoch": 2.8820362212432697,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004523181688841728,
|
|
"loss": 5.1776,
|
|
"mean_token_accuracy": 0.20025330334901809,
|
|
"num_tokens": 7009792.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.40869107246399,
|
|
"epoch": 2.886930983847283,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004520816884558925,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.19975550919771196,
|
|
"num_tokens": 7022327.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.127566289901734,
|
|
"epoch": 2.891825746451297,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045184469292465555,
|
|
"loss": 5.0348,
|
|
"mean_token_accuracy": 0.2134496822953224,
|
|
"num_tokens": 7032861.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.2581127166748045,
|
|
"epoch": 2.896720509055311,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000451607182979845,
|
|
"loss": 5.1886,
|
|
"mean_token_accuracy": 0.19754125773906708,
|
|
"num_tokens": 7045091.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.361897563934326,
|
|
"epoch": 2.9016152716593244,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004513691593123401,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.19922478795051574,
|
|
"num_tokens": 7058687.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.230517911911011,
|
|
"epoch": 2.9065100342633383,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045113062261451447,
|
|
"loss": 5.1041,
|
|
"mean_token_accuracy": 0.20409076511859894,
|
|
"num_tokens": 7070185.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.153442859649658,
|
|
"epoch": 2.9114047968673518,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004508915735802342,
|
|
"loss": 5.0436,
|
|
"mean_token_accuracy": 0.21531937271356583,
|
|
"num_tokens": 7082084.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.288640642166138,
|
|
"epoch": 2.9162995594713657,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045065201290485547,
|
|
"loss": 5.1311,
|
|
"mean_token_accuracy": 0.2090048909187317,
|
|
"num_tokens": 7093648.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.317954778671265,
|
|
"epoch": 2.9211943220753795,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004504119412852228,
|
|
"loss": 5.2205,
|
|
"mean_token_accuracy": 0.1979729115962982,
|
|
"num_tokens": 7106444.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.221889877319336,
|
|
"epoch": 2.926089084679393,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004501713594196671,
|
|
"loss": 5.1525,
|
|
"mean_token_accuracy": 0.1946215435862541,
|
|
"num_tokens": 7119287.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.262315082550049,
|
|
"epoch": 2.930983847283407,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004499302680080034,
|
|
"loss": 5.1373,
|
|
"mean_token_accuracy": 0.20100004523992537,
|
|
"num_tokens": 7130615.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.250856971740722,
|
|
"epoch": 2.9358786098874203,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000449688667751529,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.2066326454281807,
|
|
"num_tokens": 7141864.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9358786098874203,
|
|
"eval_entropy": 5.239256037913453,
|
|
"eval_loss": 5.65110969543457,
|
|
"eval_mean_token_accuracy": 0.18974790772263245,
|
|
"eval_num_tokens": 7141864.0,
|
|
"eval_runtime": 2.7121,
|
|
"eval_samples_per_second": 1270.994,
|
|
"eval_steps_per_second": 158.92,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.3498396396636965,
|
|
"epoch": 2.9407733724914342,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044944655935302114,
|
|
"loss": 5.1835,
|
|
"mean_token_accuracy": 0.19889111071825027,
|
|
"num_tokens": 7152303.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.192798328399658,
|
|
"epoch": 2.945668135095448,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004492039435167355,
|
|
"loss": 5.1174,
|
|
"mean_token_accuracy": 0.2020832344889641,
|
|
"num_tokens": 7164398.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.242094326019287,
|
|
"epoch": 2.9505628976994616,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004489608209484034,
|
|
"loss": 5.1279,
|
|
"mean_token_accuracy": 0.20991869866847992,
|
|
"num_tokens": 7175192.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.273676490783691,
|
|
"epoch": 2.955457660303475,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004487171923552306,
|
|
"loss": 5.2383,
|
|
"mean_token_accuracy": 0.19441885203123094,
|
|
"num_tokens": 7189522.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.32355146408081,
|
|
"epoch": 2.960352422907489,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004484730584458945,
|
|
"loss": 5.1586,
|
|
"mean_token_accuracy": 0.2014496371150017,
|
|
"num_tokens": 7201404.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.206380558013916,
|
|
"epoch": 2.965247185511503,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004482284199305426,
|
|
"loss": 5.1062,
|
|
"mean_token_accuracy": 0.20622824877500534,
|
|
"num_tokens": 7212355.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.183655405044556,
|
|
"epoch": 2.9701419481155162,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044798327752078993,
|
|
"loss": 5.0126,
|
|
"mean_token_accuracy": 0.21361497044563293,
|
|
"num_tokens": 7223297.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.1699216842651365,
|
|
"epoch": 2.97503671071953,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044773763192971756,
|
|
"loss": 5.0496,
|
|
"mean_token_accuracy": 0.21205516457557677,
|
|
"num_tokens": 7235019.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.302969074249267,
|
|
"epoch": 2.9799314733235436,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004474914838718701,
|
|
"loss": 5.1468,
|
|
"mean_token_accuracy": 0.2011548325419426,
|
|
"num_tokens": 7247429.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.151635313034058,
|
|
"epoch": 2.9848262359275575,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044724483406325375,
|
|
"loss": 5.0214,
|
|
"mean_token_accuracy": 0.21079832166433335,
|
|
"num_tokens": 7259211.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.201655673980713,
|
|
"epoch": 2.9897209985315714,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004469976832213344,
|
|
"loss": 5.1013,
|
|
"mean_token_accuracy": 0.20135508179664613,
|
|
"num_tokens": 7270843.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.2060614109039305,
|
|
"epoch": 2.994615761135585,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000446750032065035,
|
|
"loss": 5.0263,
|
|
"mean_token_accuracy": 0.20377332866191863,
|
|
"num_tokens": 7282644.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.1719215393066404,
|
|
"epoch": 2.9995105237395987,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004465018813147342,
|
|
"loss": 5.0764,
|
|
"mean_token_accuracy": 0.2082974672317505,
|
|
"num_tokens": 7294592.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.152208911048041,
|
|
"epoch": 3.003915810083211,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044625323169226376,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.21952111356788212,
|
|
"num_tokens": 7304315.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.258079576492309,
|
|
"epoch": 3.0088105726872247,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044600408392090653,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.2206038549542427,
|
|
"num_tokens": 7316042.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.194670152664185,
|
|
"epoch": 3.013705335291238,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004457544387253945,
|
|
"loss": 4.8032,
|
|
"mean_token_accuracy": 0.22254225313663484,
|
|
"num_tokens": 7328096.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.21949610710144,
|
|
"epoch": 3.018600097895252,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004455042968319063,
|
|
"loss": 4.8169,
|
|
"mean_token_accuracy": 0.22124014794826508,
|
|
"num_tokens": 7340093.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.261151647567749,
|
|
"epoch": 3.023494860499266,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044525365896806593,
|
|
"loss": 4.8632,
|
|
"mean_token_accuracy": 0.22004459500312806,
|
|
"num_tokens": 7352439.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.141685581207275,
|
|
"epoch": 3.0283896231032794,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044500252586293957,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.218286868929863,
|
|
"num_tokens": 7364139.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.218305540084839,
|
|
"epoch": 3.0332843857072933,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044475089824703426,
|
|
"loss": 4.8619,
|
|
"mean_token_accuracy": 0.21531160771846772,
|
|
"num_tokens": 7375390.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.173544788360596,
|
|
"epoch": 3.0381791483113068,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004444987768522952,
|
|
"loss": 4.7697,
|
|
"mean_token_accuracy": 0.2147290900349617,
|
|
"num_tokens": 7386350.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.174273061752319,
|
|
"epoch": 3.0430739109153206,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004442461624121044,
|
|
"loss": 4.7454,
|
|
"mean_token_accuracy": 0.2209487721323967,
|
|
"num_tokens": 7398256.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.181314134597779,
|
|
"epoch": 3.0479686735193345,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044399305566127773,
|
|
"loss": 4.7692,
|
|
"mean_token_accuracy": 0.219735686480999,
|
|
"num_tokens": 7410083.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.160170269012451,
|
|
"epoch": 3.052863436123348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004437394573360631,
|
|
"loss": 4.7951,
|
|
"mean_token_accuracy": 0.22411915212869643,
|
|
"num_tokens": 7421988.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.139834976196289,
|
|
"epoch": 3.057758198727362,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004434853681741384,
|
|
"loss": 4.7997,
|
|
"mean_token_accuracy": 0.21659549027681352,
|
|
"num_tokens": 7434191.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.182546091079712,
|
|
"epoch": 3.0626529613313753,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044323078891460934,
|
|
"loss": 4.7833,
|
|
"mean_token_accuracy": 0.22339957505464553,
|
|
"num_tokens": 7445765.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.123447847366333,
|
|
"epoch": 3.067547723935389,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004429757202980071,
|
|
"loss": 4.7552,
|
|
"mean_token_accuracy": 0.22224232852458953,
|
|
"num_tokens": 7457269.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.104322624206543,
|
|
"epoch": 3.0724424865394027,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044272016306628656,
|
|
"loss": 4.779,
|
|
"mean_token_accuracy": 0.22368589043617249,
|
|
"num_tokens": 7470048.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.089307832717895,
|
|
"epoch": 3.0773372491434166,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004424641179628237,
|
|
"loss": 4.7551,
|
|
"mean_token_accuracy": 0.22463481724262238,
|
|
"num_tokens": 7482954.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.12674765586853,
|
|
"epoch": 3.0822320117474304,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 4.8024,
|
|
"mean_token_accuracy": 0.22329255938529968,
|
|
"num_tokens": 7495546.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.1817950248718265,
|
|
"epoch": 3.087126774351444,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044195056712126876,
|
|
"loss": 4.8436,
|
|
"mean_token_accuracy": 0.2192997917532921,
|
|
"num_tokens": 7508042.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.086804294586182,
|
|
"epoch": 3.092021536955458,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004416930628770158,
|
|
"loss": 4.8123,
|
|
"mean_token_accuracy": 0.21833253055810928,
|
|
"num_tokens": 7520548.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.14989275932312,
|
|
"epoch": 3.0969162995594712,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044143507374869444,
|
|
"loss": 4.8172,
|
|
"mean_token_accuracy": 0.21817636638879775,
|
|
"num_tokens": 7531106.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.08776741027832,
|
|
"epoch": 3.101811062163485,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004411766004867548,
|
|
"loss": 4.7439,
|
|
"mean_token_accuracy": 0.22358861118555068,
|
|
"num_tokens": 7543159.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.1496686935424805,
|
|
"epoch": 3.1067058247674986,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044091764384305524,
|
|
"loss": 4.8313,
|
|
"mean_token_accuracy": 0.22145208567380906,
|
|
"num_tokens": 7554708.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.182804679870605,
|
|
"epoch": 3.1116005873715125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044065820457085995,
|
|
"loss": 4.8088,
|
|
"mean_token_accuracy": 0.2163847044110298,
|
|
"num_tokens": 7564896.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.093549823760986,
|
|
"epoch": 3.1164953499755264,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004403982834248374,
|
|
"loss": 4.7333,
|
|
"mean_token_accuracy": 0.2243751674890518,
|
|
"num_tokens": 7575917.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.106040048599243,
|
|
"epoch": 3.12139011257954,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044013788116105754,
|
|
"loss": 4.8104,
|
|
"mean_token_accuracy": 0.22413290143013,
|
|
"num_tokens": 7587632.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.115657091140747,
|
|
"epoch": 3.1262848751835537,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004398769985369898,
|
|
"loss": 4.6619,
|
|
"mean_token_accuracy": 0.2362922489643097,
|
|
"num_tokens": 7599177.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.15286660194397,
|
|
"epoch": 3.131179637787567,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004396156363115011,
|
|
"loss": 4.8064,
|
|
"mean_token_accuracy": 0.22309130877256395,
|
|
"num_tokens": 7611122.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.14999794960022,
|
|
"epoch": 3.136074400391581,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004393537952448531,
|
|
"loss": 4.8488,
|
|
"mean_token_accuracy": 0.21808122098445892,
|
|
"num_tokens": 7622053.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.1783332347869875,
|
|
"epoch": 3.140969162995595,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004390914760987008,
|
|
"loss": 4.8574,
|
|
"mean_token_accuracy": 0.2219322144985199,
|
|
"num_tokens": 7633774.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.17280764579773,
|
|
"epoch": 3.1458639255996084,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004388286796360895,
|
|
"loss": 4.8814,
|
|
"mean_token_accuracy": 0.21389546245336533,
|
|
"num_tokens": 7645286.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.111659717559815,
|
|
"epoch": 3.1507586882036223,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004385654066214529,
|
|
"loss": 4.7673,
|
|
"mean_token_accuracy": 0.22385075837373733,
|
|
"num_tokens": 7656576.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.107363080978393,
|
|
"epoch": 3.1556534508076357,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004383016578206112,
|
|
"loss": 4.8201,
|
|
"mean_token_accuracy": 0.22123815417289733,
|
|
"num_tokens": 7667786.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.20523624420166,
|
|
"epoch": 3.1605482134116496,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004380374340007684,
|
|
"loss": 4.8452,
|
|
"mean_token_accuracy": 0.2151228219270706,
|
|
"num_tokens": 7680029.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.136455726623535,
|
|
"epoch": 3.165442976015663,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004377727359305103,
|
|
"loss": 4.8486,
|
|
"mean_token_accuracy": 0.2207125499844551,
|
|
"num_tokens": 7692659.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.150415134429932,
|
|
"epoch": 3.170337738619677,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004375075643798022,
|
|
"loss": 4.8128,
|
|
"mean_token_accuracy": 0.2215397596359253,
|
|
"num_tokens": 7705011.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.054905748367309,
|
|
"epoch": 3.175232501223691,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004372419201199867,
|
|
"loss": 4.7561,
|
|
"mean_token_accuracy": 0.22285788357257844,
|
|
"num_tokens": 7717489.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.087309694290161,
|
|
"epoch": 3.1801272638277043,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004369758039237814,
|
|
"loss": 4.7968,
|
|
"mean_token_accuracy": 0.22163646519184113,
|
|
"num_tokens": 7730009.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.211210298538208,
|
|
"epoch": 3.185022026431718,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043670921656527686,
|
|
"loss": 4.8711,
|
|
"mean_token_accuracy": 0.21754691898822784,
|
|
"num_tokens": 7741782.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.170057535171509,
|
|
"epoch": 3.1899167890357316,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000436442158819934,
|
|
"loss": 4.8354,
|
|
"mean_token_accuracy": 0.21667892038822173,
|
|
"num_tokens": 7753135.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.098474073410034,
|
|
"epoch": 3.1948115516397455,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043617463146458196,
|
|
"loss": 4.8281,
|
|
"mean_token_accuracy": 0.21399773508310319,
|
|
"num_tokens": 7766344.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.120054292678833,
|
|
"epoch": 3.199706314243759,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000435906635277416,
|
|
"loss": 4.8442,
|
|
"mean_token_accuracy": 0.2192484274506569,
|
|
"num_tokens": 7778419.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.065897846221924,
|
|
"epoch": 3.204601076847773,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004356381710379954,
|
|
"loss": 4.7307,
|
|
"mean_token_accuracy": 0.2220268577337265,
|
|
"num_tokens": 7789303.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.102967166900635,
|
|
"epoch": 3.2094958394517867,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004353692395272404,
|
|
"loss": 4.8441,
|
|
"mean_token_accuracy": 0.2154310703277588,
|
|
"num_tokens": 7803315.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.133005380630493,
|
|
"epoch": 3.2143906020558,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043509984152743074,
|
|
"loss": 4.8317,
|
|
"mean_token_accuracy": 0.21344627887010575,
|
|
"num_tokens": 7814651.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.086959648132324,
|
|
"epoch": 3.219285364659814,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043482997782220333,
|
|
"loss": 4.7976,
|
|
"mean_token_accuracy": 0.21823935210704803,
|
|
"num_tokens": 7826481.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.1153992176055905,
|
|
"epoch": 3.2241801272638275,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043455964919654927,
|
|
"loss": 4.7424,
|
|
"mean_token_accuracy": 0.23253000527620316,
|
|
"num_tokens": 7837868.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.099810552597046,
|
|
"epoch": 3.2290748898678414,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004342888564368124,
|
|
"loss": 4.8011,
|
|
"mean_token_accuracy": 0.22438711524009705,
|
|
"num_tokens": 7850928.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.163304328918457,
|
|
"epoch": 3.2339696524718553,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043401760033068656,
|
|
"loss": 4.8854,
|
|
"mean_token_accuracy": 0.2155757039785385,
|
|
"num_tokens": 7862343.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.148097896575928,
|
|
"epoch": 3.2388644150758688,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004337458816672132,
|
|
"loss": 4.8496,
|
|
"mean_token_accuracy": 0.21238613426685332,
|
|
"num_tokens": 7874681.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.073517274856568,
|
|
"epoch": 3.2437591776798826,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004334737012367797,
|
|
"loss": 4.7769,
|
|
"mean_token_accuracy": 0.21962419748306275,
|
|
"num_tokens": 7886624.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.226518678665161,
|
|
"epoch": 3.248653940283896,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043320105983111627,
|
|
"loss": 4.8749,
|
|
"mean_token_accuracy": 0.21509021818637847,
|
|
"num_tokens": 7897918.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.152651309967041,
|
|
"epoch": 3.25354870288791,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004329279582432941,
|
|
"loss": 4.932,
|
|
"mean_token_accuracy": 0.21958141177892684,
|
|
"num_tokens": 7909698.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.121517133712769,
|
|
"epoch": 3.258443465491924,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004326543972677231,
|
|
"loss": 4.8369,
|
|
"mean_token_accuracy": 0.21473742574453353,
|
|
"num_tokens": 7921715.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.076841068267822,
|
|
"epoch": 3.2633382280959373,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043238037770014937,
|
|
"loss": 4.7511,
|
|
"mean_token_accuracy": 0.21911688148975372,
|
|
"num_tokens": 7933262.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.141078281402588,
|
|
"epoch": 3.268232990699951,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000432105900337653,
|
|
"loss": 4.8511,
|
|
"mean_token_accuracy": 0.21800520569086074,
|
|
"num_tokens": 7944352.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.0780316352844235,
|
|
"epoch": 3.2731277533039647,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004318309659786457,
|
|
"loss": 4.7584,
|
|
"mean_token_accuracy": 0.22223492413759233,
|
|
"num_tokens": 7955024.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.087853527069091,
|
|
"epoch": 3.2780225159079786,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004315555754228687,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.22424828857183457,
|
|
"num_tokens": 7965543.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.142214775085449,
|
|
"epoch": 3.282917278511992,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043127972947138994,
|
|
"loss": 4.8638,
|
|
"mean_token_accuracy": 0.2150084137916565,
|
|
"num_tokens": 7977179.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.203800439834595,
|
|
"epoch": 3.287812041116006,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043100342892660225,
|
|
"loss": 4.9112,
|
|
"mean_token_accuracy": 0.20913542062044144,
|
|
"num_tokens": 7990177.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.177821969985962,
|
|
"epoch": 3.2927068037200193,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004307266745922207,
|
|
"loss": 4.7909,
|
|
"mean_token_accuracy": 0.2120165690779686,
|
|
"num_tokens": 8000879.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.078297185897827,
|
|
"epoch": 3.2976015663240332,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043044946727328036,
|
|
"loss": 4.8003,
|
|
"mean_token_accuracy": 0.21971254646778107,
|
|
"num_tokens": 8013335.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.103021907806396,
|
|
"epoch": 3.302496328928047,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043017180777613415,
|
|
"loss": 4.8135,
|
|
"mean_token_accuracy": 0.22505059242248535,
|
|
"num_tokens": 8025020.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.110682249069214,
|
|
"epoch": 3.3073910915320606,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042989369690844993,
|
|
"loss": 4.7648,
|
|
"mean_token_accuracy": 0.2246474251151085,
|
|
"num_tokens": 8037244.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.001568698883057,
|
|
"epoch": 3.3122858541360745,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000429615135479209,
|
|
"loss": 4.7959,
|
|
"mean_token_accuracy": 0.22334180027246475,
|
|
"num_tokens": 8049087.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.188577890396118,
|
|
"epoch": 3.317180616740088,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042933612429870284,
|
|
"loss": 4.9156,
|
|
"mean_token_accuracy": 0.21281692385673523,
|
|
"num_tokens": 8061052.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.12839412689209,
|
|
"epoch": 3.322075379344102,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004290566641785314,
|
|
"loss": 4.7894,
|
|
"mean_token_accuracy": 0.22035609632730485,
|
|
"num_tokens": 8072389.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.003451681137085,
|
|
"epoch": 3.3269701419481157,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042877675593160053,
|
|
"loss": 4.8,
|
|
"mean_token_accuracy": 0.21525861769914628,
|
|
"num_tokens": 8085141.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.1706421852111815,
|
|
"epoch": 3.331864904552129,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042849640037211957,
|
|
"loss": 4.9041,
|
|
"mean_token_accuracy": 0.2083066239953041,
|
|
"num_tokens": 8097337.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.114046382904053,
|
|
"epoch": 3.336759667156143,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000428215598315599,
|
|
"loss": 4.8412,
|
|
"mean_token_accuracy": 0.21854109466075897,
|
|
"num_tokens": 8109160.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.143175458908081,
|
|
"epoch": 3.3416544297601565,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042793435057884816,
|
|
"loss": 4.7823,
|
|
"mean_token_accuracy": 0.21902110427618027,
|
|
"num_tokens": 8121195.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.103787803649903,
|
|
"epoch": 3.3465491923641704,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042765265797997274,
|
|
"loss": 4.7733,
|
|
"mean_token_accuracy": 0.22462240755558013,
|
|
"num_tokens": 8132364.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.122178840637207,
|
|
"epoch": 3.3514439549681843,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042737052133837253,
|
|
"loss": 4.8555,
|
|
"mean_token_accuracy": 0.21716293841600418,
|
|
"num_tokens": 8144703.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.119469022750854,
|
|
"epoch": 3.3563387175721977,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042708794147473876,
|
|
"loss": 4.8393,
|
|
"mean_token_accuracy": 0.21995826959609985,
|
|
"num_tokens": 8157364.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.194144201278687,
|
|
"epoch": 3.3612334801762116,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004268049192110523,
|
|
"loss": 4.9333,
|
|
"mean_token_accuracy": 0.21170254945755004,
|
|
"num_tokens": 8170264.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.169523477554321,
|
|
"epoch": 3.366128242780225,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042652145537058056,
|
|
"loss": 4.865,
|
|
"mean_token_accuracy": 0.2144750773906708,
|
|
"num_tokens": 8182854.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.071971988677978,
|
|
"epoch": 3.371023005384239,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042623755077787556,
|
|
"loss": 4.8302,
|
|
"mean_token_accuracy": 0.21609048843383788,
|
|
"num_tokens": 8195098.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.1184186935424805,
|
|
"epoch": 3.3759177679882524,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042595320625877153,
|
|
"loss": 4.9155,
|
|
"mean_token_accuracy": 0.21652448922395706,
|
|
"num_tokens": 8207513.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.167665195465088,
|
|
"epoch": 3.3808125305922663,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00042566842264038203,
|
|
"loss": 4.8084,
|
|
"mean_token_accuracy": 0.2188054382801056,
|
|
"num_tokens": 8218603.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.169710397720337,
|
|
"epoch": 3.38570729319628,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004253832007510982,
|
|
"loss": 4.8944,
|
|
"mean_token_accuracy": 0.21486698538064958,
|
|
"num_tokens": 8231624.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.103657531738281,
|
|
"epoch": 3.3906020558002936,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000425097541420586,
|
|
"loss": 4.8423,
|
|
"mean_token_accuracy": 0.21821292191743852,
|
|
"num_tokens": 8242119.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.117683172225952,
|
|
"epoch": 3.3954968184043075,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004248114454797837,
|
|
"loss": 4.807,
|
|
"mean_token_accuracy": 0.2168683558702469,
|
|
"num_tokens": 8252944.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.120380592346192,
|
|
"epoch": 3.400391581008321,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042452491376089964,
|
|
"loss": 4.893,
|
|
"mean_token_accuracy": 0.21788542568683625,
|
|
"num_tokens": 8264761.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.08057017326355,
|
|
"epoch": 3.405286343612335,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004242379470974099,
|
|
"loss": 4.8175,
|
|
"mean_token_accuracy": 0.22083471715450287,
|
|
"num_tokens": 8278172.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.121869325637817,
|
|
"epoch": 3.4101811062163483,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042395054632405563,
|
|
"loss": 4.8658,
|
|
"mean_token_accuracy": 0.21356443911790848,
|
|
"num_tokens": 8290836.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.146685361862183,
|
|
"epoch": 3.415075868820362,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004236627122768406,
|
|
"loss": 4.8467,
|
|
"mean_token_accuracy": 0.22179711163043975,
|
|
"num_tokens": 8302467.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.041498041152954,
|
|
"epoch": 3.419970631424376,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004233744457930294,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.2170724630355835,
|
|
"num_tokens": 8315190.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.161109066009521,
|
|
"epoch": 3.4248653940283895,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042308574771114394,
|
|
"loss": 4.9148,
|
|
"mean_token_accuracy": 0.21417440623044967,
|
|
"num_tokens": 8327463.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.111668586730957,
|
|
"epoch": 3.4297601566324034,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004227966188709618,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.21845465749502183,
|
|
"num_tokens": 8339005.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.139991712570191,
|
|
"epoch": 3.434654919236417,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004225070601135138,
|
|
"loss": 4.9125,
|
|
"mean_token_accuracy": 0.21437896490097047,
|
|
"num_tokens": 8351478.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.168607521057129,
|
|
"epoch": 3.4395496818404308,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000422217072281081,
|
|
"loss": 4.8433,
|
|
"mean_token_accuracy": 0.21583523452281952,
|
|
"num_tokens": 8363939.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.17737832069397,
|
|
"epoch": 3.4444444444444446,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004219266562171925,
|
|
"loss": 5.0011,
|
|
"mean_token_accuracy": 0.2089764431118965,
|
|
"num_tokens": 8377086.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.040504741668701,
|
|
"epoch": 3.449339207048458,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004216358127666235,
|
|
"loss": 4.7402,
|
|
"mean_token_accuracy": 0.2260405331850052,
|
|
"num_tokens": 8389085.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.103902959823609,
|
|
"epoch": 3.454233969652472,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042134454277539205,
|
|
"loss": 4.829,
|
|
"mean_token_accuracy": 0.21698508113622667,
|
|
"num_tokens": 8401240.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 4.998382949829102,
|
|
"epoch": 3.4591287322564854,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000421052847090757,
|
|
"loss": 4.7466,
|
|
"mean_token_accuracy": 0.2224629834294319,
|
|
"num_tokens": 8413988.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.103152704238892,
|
|
"epoch": 3.4640234948604993,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004207607265612156,
|
|
"loss": 4.8352,
|
|
"mean_token_accuracy": 0.2225386008620262,
|
|
"num_tokens": 8425106.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.090368127822876,
|
|
"epoch": 3.468918257464513,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042046818203650074,
|
|
"loss": 4.86,
|
|
"mean_token_accuracy": 0.2257356509566307,
|
|
"num_tokens": 8437232.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.079847431182861,
|
|
"epoch": 3.4738130200685267,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042017521436757876,
|
|
"loss": 4.8133,
|
|
"mean_token_accuracy": 0.22792982906103135,
|
|
"num_tokens": 8448403.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.122736120223999,
|
|
"epoch": 3.4787077826725405,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004198818244066468,
|
|
"loss": 4.8588,
|
|
"mean_token_accuracy": 0.21321402192115785,
|
|
"num_tokens": 8461044.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.096948671340942,
|
|
"epoch": 3.483602545276554,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041958801300713035,
|
|
"loss": 4.8599,
|
|
"mean_token_accuracy": 0.21646499633789062,
|
|
"num_tokens": 8473829.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.096843147277832,
|
|
"epoch": 3.488497307880568,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000419293781023681,
|
|
"loss": 4.8512,
|
|
"mean_token_accuracy": 0.22051677703857422,
|
|
"num_tokens": 8485033.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.0834801197052,
|
|
"epoch": 3.4933920704845813,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041899912931217347,
|
|
"loss": 4.8231,
|
|
"mean_token_accuracy": 0.22450067847967148,
|
|
"num_tokens": 8498307.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.0844282627105715,
|
|
"epoch": 3.4982868330885952,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041870405872970355,
|
|
"loss": 4.9073,
|
|
"mean_token_accuracy": 0.21928595751523972,
|
|
"num_tokens": 8510766.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.100914096832275,
|
|
"epoch": 3.5031815956926087,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004184085701345856,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.2184406891465187,
|
|
"num_tokens": 8523041.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.121054792404175,
|
|
"epoch": 3.5080763582966226,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004181126643863496,
|
|
"loss": 4.8354,
|
|
"mean_token_accuracy": 0.2177764892578125,
|
|
"num_tokens": 8535984.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.093640518188477,
|
|
"epoch": 3.5129711209006365,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004178163423457391,
|
|
"loss": 4.8075,
|
|
"mean_token_accuracy": 0.22597914487123488,
|
|
"num_tokens": 8548061.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.09451470375061,
|
|
"epoch": 3.51786588350465,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041751960487470884,
|
|
"loss": 4.7958,
|
|
"mean_token_accuracy": 0.22748810350894927,
|
|
"num_tokens": 8558145.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.029383277893066,
|
|
"epoch": 3.522760646108664,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004172224528364215,
|
|
"loss": 4.7723,
|
|
"mean_token_accuracy": 0.2199097290635109,
|
|
"num_tokens": 8570007.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.113267087936402,
|
|
"epoch": 3.5276554087126772,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000416924887095246,
|
|
"loss": 4.8472,
|
|
"mean_token_accuracy": 0.2192206010222435,
|
|
"num_tokens": 8582162.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.056365060806274,
|
|
"epoch": 3.532550171316691,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041662690851675463,
|
|
"loss": 4.7057,
|
|
"mean_token_accuracy": 0.22681989520788193,
|
|
"num_tokens": 8593205.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.028490972518921,
|
|
"epoch": 3.537444933920705,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004163285179677203,
|
|
"loss": 4.7384,
|
|
"mean_token_accuracy": 0.22807639688253403,
|
|
"num_tokens": 8604576.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.0321262836456295,
|
|
"epoch": 3.5423396965247185,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004160297163161148,
|
|
"loss": 4.8157,
|
|
"mean_token_accuracy": 0.21812277138233185,
|
|
"num_tokens": 8617252.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 4.983499479293823,
|
|
"epoch": 3.5472344591287324,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004157305044311053,
|
|
"loss": 4.7504,
|
|
"mean_token_accuracy": 0.22493973821401597,
|
|
"num_tokens": 8627817.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.115007448196411,
|
|
"epoch": 3.552129221732746,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041543088318305235,
|
|
"loss": 4.8982,
|
|
"mean_token_accuracy": 0.21084729284048082,
|
|
"num_tokens": 8640246.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.066983366012574,
|
|
"epoch": 3.5570239843367597,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004151308534435073,
|
|
"loss": 4.8294,
|
|
"mean_token_accuracy": 0.21993755400180817,
|
|
"num_tokens": 8651695.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.1451818466186525,
|
|
"epoch": 3.5619187469407736,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041483041608521,
|
|
"loss": 4.8984,
|
|
"mean_token_accuracy": 0.21316303610801696,
|
|
"num_tokens": 8664529.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.082834625244141,
|
|
"epoch": 3.566813509544787,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004145295719820855,
|
|
"loss": 4.845,
|
|
"mean_token_accuracy": 0.21952054500579835,
|
|
"num_tokens": 8676203.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.060942316055298,
|
|
"epoch": 3.571708272148801,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004142283220092425,
|
|
"loss": 4.8212,
|
|
"mean_token_accuracy": 0.22039462327957154,
|
|
"num_tokens": 8688790.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.050752830505371,
|
|
"epoch": 3.5766030347528144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041392666704296986,
|
|
"loss": 4.8507,
|
|
"mean_token_accuracy": 0.2112608879804611,
|
|
"num_tokens": 8699179.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.032579851150513,
|
|
"epoch": 3.5814977973568283,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041362460796073495,
|
|
"loss": 4.742,
|
|
"mean_token_accuracy": 0.22416270524263382,
|
|
"num_tokens": 8710883.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.108354282379151,
|
|
"epoch": 3.586392559960842,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004133221456411803,
|
|
"loss": 4.9056,
|
|
"mean_token_accuracy": 0.21302346438169478,
|
|
"num_tokens": 8721425.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.165343856811523,
|
|
"epoch": 3.5912873225648556,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041301928096412164,
|
|
"loss": 4.8661,
|
|
"mean_token_accuracy": 0.21713459640741348,
|
|
"num_tokens": 8733809.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.068774461746216,
|
|
"epoch": 3.596182085168869,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041271601481054486,
|
|
"loss": 4.782,
|
|
"mean_token_accuracy": 0.22668402045965194,
|
|
"num_tokens": 8746370.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.073219871520996,
|
|
"epoch": 3.601076847772883,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004124123480626039,
|
|
"loss": 4.9065,
|
|
"mean_token_accuracy": 0.2157248854637146,
|
|
"num_tokens": 8758655.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.111428451538086,
|
|
"epoch": 3.605971610376897,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004121082816036178,
|
|
"loss": 4.8241,
|
|
"mean_token_accuracy": 0.22177139967679976,
|
|
"num_tokens": 8770171.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.085653305053711,
|
|
"epoch": 3.6108663729809103,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004118038163180684,
|
|
"loss": 4.8002,
|
|
"mean_token_accuracy": 0.22371812760829926,
|
|
"num_tokens": 8783015.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.092598009109497,
|
|
"epoch": 3.615761135584924,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041149895309159756,
|
|
"loss": 4.8909,
|
|
"mean_token_accuracy": 0.20332240909337998,
|
|
"num_tokens": 8795382.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.08798360824585,
|
|
"epoch": 3.6206558981889376,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000411193692811005,
|
|
"loss": 4.7672,
|
|
"mean_token_accuracy": 0.22842831909656525,
|
|
"num_tokens": 8806634.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.099050235748291,
|
|
"epoch": 3.6255506607929515,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041088803636424494,
|
|
"loss": 4.8453,
|
|
"mean_token_accuracy": 0.2124974772334099,
|
|
"num_tokens": 8818926.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.049224233627319,
|
|
"epoch": 3.6304454233969654,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041058198464042424,
|
|
"loss": 4.8035,
|
|
"mean_token_accuracy": 0.2252272590994835,
|
|
"num_tokens": 8830267.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.075808572769165,
|
|
"epoch": 3.635340186000979,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004102755385297996,
|
|
"loss": 4.8897,
|
|
"mean_token_accuracy": 0.21413865089416503,
|
|
"num_tokens": 8843330.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.051680183410644,
|
|
"epoch": 3.6402349486049927,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040996869892377473,
|
|
"loss": 4.8087,
|
|
"mean_token_accuracy": 0.22400161176919936,
|
|
"num_tokens": 8853467.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.0674622535705565,
|
|
"epoch": 3.645129711209006,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040966146671489836,
|
|
"loss": 4.8341,
|
|
"mean_token_accuracy": 0.21511287689208985,
|
|
"num_tokens": 8868203.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.044716215133667,
|
|
"epoch": 3.65002447381302,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040935384279686067,
|
|
"loss": 4.7361,
|
|
"mean_token_accuracy": 0.22642216980457305,
|
|
"num_tokens": 8879171.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 4.989019775390625,
|
|
"epoch": 3.654919236417034,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040904582806449173,
|
|
"loss": 4.723,
|
|
"mean_token_accuracy": 0.22874550074338912,
|
|
"num_tokens": 8890459.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.059218406677246,
|
|
"epoch": 3.6598139990210474,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040873742341375826,
|
|
"loss": 4.8125,
|
|
"mean_token_accuracy": 0.22379768639802933,
|
|
"num_tokens": 8902878.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.009291505813598,
|
|
"epoch": 3.6647087616250613,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00040842862974176114,
|
|
"loss": 4.7337,
|
|
"mean_token_accuracy": 0.22724575400352479,
|
|
"num_tokens": 8914792.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 4.9672339916229244,
|
|
"epoch": 3.6696035242290748,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004081194479467329,
|
|
"loss": 4.75,
|
|
"mean_token_accuracy": 0.22816742211580276,
|
|
"num_tokens": 8926610.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.06518964767456,
|
|
"epoch": 3.6744982868330887,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00040780987892803514,
|
|
"loss": 4.8625,
|
|
"mean_token_accuracy": 0.22146295756101608,
|
|
"num_tokens": 8940250.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.131087017059326,
|
|
"epoch": 3.6793930494371025,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040749992358615555,
|
|
"loss": 4.8046,
|
|
"mean_token_accuracy": 0.2123077243566513,
|
|
"num_tokens": 8951960.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 4.953831195831299,
|
|
"epoch": 3.684287812041116,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000407189582822706,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.2197168067097664,
|
|
"num_tokens": 8963265.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.039992570877075,
|
|
"epoch": 3.6891825746451294,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004068788575404192,
|
|
"loss": 4.8209,
|
|
"mean_token_accuracy": 0.22028125673532487,
|
|
"num_tokens": 8974679.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.09268364906311,
|
|
"epoch": 3.6940773372491433,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004065677486431464,
|
|
"loss": 4.8292,
|
|
"mean_token_accuracy": 0.22199592888355255,
|
|
"num_tokens": 8986840.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.113239908218384,
|
|
"epoch": 3.6989720998531572,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004062562570358548,
|
|
"loss": 4.8443,
|
|
"mean_token_accuracy": 0.21674781739711763,
|
|
"num_tokens": 8997926.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.064158153533936,
|
|
"epoch": 3.7038668624571707,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040594438362462484,
|
|
"loss": 4.8083,
|
|
"mean_token_accuracy": 0.2272493675351143,
|
|
"num_tokens": 9009668.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.071455669403076,
|
|
"epoch": 3.7087616250611846,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004056321293166474,
|
|
"loss": 4.8264,
|
|
"mean_token_accuracy": 0.2173181027173996,
|
|
"num_tokens": 9021611.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.07202558517456,
|
|
"epoch": 3.713656387665198,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040531949502022155,
|
|
"loss": 4.803,
|
|
"mean_token_accuracy": 0.22555143237113953,
|
|
"num_tokens": 9033258.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.166383552551269,
|
|
"epoch": 3.718551150269212,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040500648164475155,
|
|
"loss": 4.9353,
|
|
"mean_token_accuracy": 0.21520323008298875,
|
|
"num_tokens": 9045264.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.069120931625366,
|
|
"epoch": 3.723445912873226,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004046930901007445,
|
|
"loss": 4.8851,
|
|
"mean_token_accuracy": 0.21698033958673477,
|
|
"num_tokens": 9058072.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.129068422317505,
|
|
"epoch": 3.7283406754772392,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004043793212998073,
|
|
"loss": 4.8574,
|
|
"mean_token_accuracy": 0.22141923755407333,
|
|
"num_tokens": 9070982.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.083976364135742,
|
|
"epoch": 3.733235438081253,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004040651761546445,
|
|
"loss": 4.9035,
|
|
"mean_token_accuracy": 0.21738988012075425,
|
|
"num_tokens": 9083520.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.115589666366577,
|
|
"epoch": 3.7381302006852666,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004037506555790551,
|
|
"loss": 4.9049,
|
|
"mean_token_accuracy": 0.21713944375514985,
|
|
"num_tokens": 9095729.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.086327981948853,
|
|
"epoch": 3.7430249632892805,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004034357604879303,
|
|
"loss": 4.8131,
|
|
"mean_token_accuracy": 0.2237031042575836,
|
|
"num_tokens": 9107400.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.108111715316772,
|
|
"epoch": 3.7479197258932944,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004031204917972508,
|
|
"loss": 4.84,
|
|
"mean_token_accuracy": 0.21787605285644532,
|
|
"num_tokens": 9119557.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.075828647613525,
|
|
"epoch": 3.752814488497308,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040280485042408384,
|
|
"loss": 4.836,
|
|
"mean_token_accuracy": 0.222735133767128,
|
|
"num_tokens": 9131862.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.147574329376221,
|
|
"epoch": 3.7577092511013217,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040248883728658094,
|
|
"loss": 4.8571,
|
|
"mean_token_accuracy": 0.2193186029791832,
|
|
"num_tokens": 9143250.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.129842519760132,
|
|
"epoch": 3.762604013705335,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004021724533039748,
|
|
"loss": 4.9094,
|
|
"mean_token_accuracy": 0.2181650459766388,
|
|
"num_tokens": 9155565.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 4.93073034286499,
|
|
"epoch": 3.767498776309349,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040185569939657717,
|
|
"loss": 4.7068,
|
|
"mean_token_accuracy": 0.2270694926381111,
|
|
"num_tokens": 9166186.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.081850004196167,
|
|
"epoch": 3.772393538913363,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040153857648577557,
|
|
"loss": 4.8421,
|
|
"mean_token_accuracy": 0.22167644798755645,
|
|
"num_tokens": 9178756.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.139322662353516,
|
|
"epoch": 3.7772883015173764,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004012210854940309,
|
|
"loss": 4.9906,
|
|
"mean_token_accuracy": 0.21866356134414672,
|
|
"num_tokens": 9190613.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.099810838699341,
|
|
"epoch": 3.7821830641213903,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000400903227344875,
|
|
"loss": 4.9191,
|
|
"mean_token_accuracy": 0.21503413170576097,
|
|
"num_tokens": 9203255.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.059392738342285,
|
|
"epoch": 3.7870778267254037,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004005850029629074,
|
|
"loss": 4.8367,
|
|
"mean_token_accuracy": 0.21643659919500352,
|
|
"num_tokens": 9215512.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.122841548919678,
|
|
"epoch": 3.7919725893294176,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004002664132737932,
|
|
"loss": 4.8445,
|
|
"mean_token_accuracy": 0.2175495907664299,
|
|
"num_tokens": 9226888.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.131028938293457,
|
|
"epoch": 3.7968673519334315,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039994745920425996,
|
|
"loss": 4.8552,
|
|
"mean_token_accuracy": 0.22309367805719377,
|
|
"num_tokens": 9238658.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.046094560623169,
|
|
"epoch": 3.801762114537445,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003996281416820952,
|
|
"loss": 4.761,
|
|
"mean_token_accuracy": 0.23520938456058502,
|
|
"num_tokens": 9249453.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.060846567153931,
|
|
"epoch": 3.8066568771414584,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003993084616361438,
|
|
"loss": 4.7883,
|
|
"mean_token_accuracy": 0.21691110581159592,
|
|
"num_tokens": 9260648.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.0653399467468265,
|
|
"epoch": 3.8115516397454723,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003989884199963049,
|
|
"loss": 4.8212,
|
|
"mean_token_accuracy": 0.218986214697361,
|
|
"num_tokens": 9273120.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.0901384353637695,
|
|
"epoch": 3.816446402349486,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003986680176935297,
|
|
"loss": 4.8453,
|
|
"mean_token_accuracy": 0.22201039493083954,
|
|
"num_tokens": 9285460.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.052167367935181,
|
|
"epoch": 3.8213411649534996,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003983472556598184,
|
|
"loss": 4.8365,
|
|
"mean_token_accuracy": 0.2265273168683052,
|
|
"num_tokens": 9297038.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.058988428115844,
|
|
"epoch": 3.8262359275575135,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003980261348282177,
|
|
"loss": 4.7678,
|
|
"mean_token_accuracy": 0.23172608464956285,
|
|
"num_tokens": 9309006.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.029935455322265,
|
|
"epoch": 3.831130690161527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003977046561328178,
|
|
"loss": 4.7527,
|
|
"mean_token_accuracy": 0.23079726994037628,
|
|
"num_tokens": 9321276.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 4.97658486366272,
|
|
"epoch": 3.836025452765541,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039738282050875,
|
|
"loss": 4.7715,
|
|
"mean_token_accuracy": 0.21992994844913483,
|
|
"num_tokens": 9333172.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.103640079498291,
|
|
"epoch": 3.8409202153695547,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003970606288921839,
|
|
"loss": 4.8113,
|
|
"mean_token_accuracy": 0.22476234287023544,
|
|
"num_tokens": 9343653.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.034993600845337,
|
|
"epoch": 3.845814977973568,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039673808222032457,
|
|
"loss": 4.8093,
|
|
"mean_token_accuracy": 0.22278518229722977,
|
|
"num_tokens": 9355811.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 4.919755554199218,
|
|
"epoch": 3.850709740577582,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003964151814314099,
|
|
"loss": 4.7726,
|
|
"mean_token_accuracy": 0.22335645258426667,
|
|
"num_tokens": 9367132.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.017191553115845,
|
|
"epoch": 3.8556045031815955,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003960919274647075,
|
|
"loss": 4.7673,
|
|
"mean_token_accuracy": 0.22698212265968323,
|
|
"num_tokens": 9379101.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.057470798492432,
|
|
"epoch": 3.8604992657856094,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000395768321260513,
|
|
"loss": 4.8288,
|
|
"mean_token_accuracy": 0.22712221890687942,
|
|
"num_tokens": 9390270.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.064850902557373,
|
|
"epoch": 3.8653940283896233,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003954443637601462,
|
|
"loss": 4.842,
|
|
"mean_token_accuracy": 0.2169718399643898,
|
|
"num_tokens": 9401724.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.049954080581665,
|
|
"epoch": 3.8702887909936368,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039512005590594876,
|
|
"loss": 4.8423,
|
|
"mean_token_accuracy": 0.2201844722032547,
|
|
"num_tokens": 9415206.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.194021081924438,
|
|
"epoch": 3.8751835535976507,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039479539864128164,
|
|
"loss": 4.8965,
|
|
"mean_token_accuracy": 0.21629704385995865,
|
|
"num_tokens": 9425830.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 4.963757419586182,
|
|
"epoch": 3.880078316201664,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039447039291052194,
|
|
"loss": 4.7876,
|
|
"mean_token_accuracy": 0.22419182658195497,
|
|
"num_tokens": 9437603.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 4.988585805892944,
|
|
"epoch": 3.884973078805678,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039414503965906075,
|
|
"loss": 4.8142,
|
|
"mean_token_accuracy": 0.22689663469791413,
|
|
"num_tokens": 9448138.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.16522707939148,
|
|
"epoch": 3.889867841409692,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003938193398332995,
|
|
"loss": 4.8378,
|
|
"mean_token_accuracy": 0.2153660163283348,
|
|
"num_tokens": 9460530.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.092146348953247,
|
|
"epoch": 3.8947626040137053,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039349329438064835,
|
|
"loss": 4.8541,
|
|
"mean_token_accuracy": 0.2215948611497879,
|
|
"num_tokens": 9472242.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.041393804550171,
|
|
"epoch": 3.899657366617719,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003931669042495226,
|
|
"loss": 4.7491,
|
|
"mean_token_accuracy": 0.22804808169603347,
|
|
"num_tokens": 9483266.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.045307779312134,
|
|
"epoch": 3.9045521292217327,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039284017038933994,
|
|
"loss": 4.7909,
|
|
"mean_token_accuracy": 0.22371412813663483,
|
|
"num_tokens": 9494032.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.0007702827453615,
|
|
"epoch": 3.9094468918257466,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039251309375051824,
|
|
"loss": 4.7481,
|
|
"mean_token_accuracy": 0.23098208010196686,
|
|
"num_tokens": 9506301.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.028079414367676,
|
|
"epoch": 3.91434165442976,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.00039218567528447225,
|
|
"loss": 4.8206,
|
|
"mean_token_accuracy": 0.22137255370616912,
|
|
"num_tokens": 9517972.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.91434165442976,
|
|
"eval_entropy": 4.977437346827956,
|
|
"eval_loss": 5.587992191314697,
|
|
"eval_mean_token_accuracy": 0.19564044451353998,
|
|
"eval_num_tokens": 9517972.0,
|
|
"eval_runtime": 2.724,
|
|
"eval_samples_per_second": 1265.405,
|
|
"eval_steps_per_second": 158.222,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10210,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.4762615014656e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|