21266 lines
582 KiB
JSON
21266 lines
582 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.020627417275461,
|
|
"eval_steps": 500,
|
|
"global_step": 10500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.69186782836914,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 14.0625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.767,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 8559.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691853904724121,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 13.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7343,
|
|
"mean_token_accuracy": 0.0006453584705013782,
|
|
"num_tokens": 18235.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.68972110748291,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 10.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4481,
|
|
"mean_token_accuracy": 0.04634292460978031,
|
|
"num_tokens": 28257.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.653952026367188,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 6.28125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1016,
|
|
"mean_token_accuracy": 0.05583016015589237,
|
|
"num_tokens": 38142.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.492306423187255,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 3.90625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8388,
|
|
"mean_token_accuracy": 0.051388037577271464,
|
|
"num_tokens": 47363.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.393956565856934,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 3.09375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.6855,
|
|
"mean_token_accuracy": 0.05032752677798271,
|
|
"num_tokens": 57121.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.352927017211915,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.59,
|
|
"mean_token_accuracy": 0.056775762140750884,
|
|
"num_tokens": 66785.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.365454959869385,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.4974,
|
|
"mean_token_accuracy": 0.05595241449773312,
|
|
"num_tokens": 76313.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.268372726440429,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.546,
|
|
"mean_token_accuracy": 0.05818333774805069,
|
|
"num_tokens": 86151.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.402025985717774,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4872,
|
|
"mean_token_accuracy": 0.06535410135984421,
|
|
"num_tokens": 94916.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.304343509674073,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4098,
|
|
"mean_token_accuracy": 0.06619026586413383,
|
|
"num_tokens": 104592.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.089207649230957,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.317,
|
|
"mean_token_accuracy": 0.07636640071868897,
|
|
"num_tokens": 113942.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.178072547912597,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.329,
|
|
"mean_token_accuracy": 0.07130967006087303,
|
|
"num_tokens": 123190.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.223089408874511,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3231,
|
|
"mean_token_accuracy": 0.06454960964620113,
|
|
"num_tokens": 132367.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.157601451873779,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1389,
|
|
"mean_token_accuracy": 0.07756163887679576,
|
|
"num_tokens": 141375.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.095236587524415,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0625,
|
|
"mean_token_accuracy": 0.07905065715312957,
|
|
"num_tokens": 150791.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.089058017730713,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0294,
|
|
"mean_token_accuracy": 0.07317423708736896,
|
|
"num_tokens": 158686.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.112849235534668,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9757,
|
|
"mean_token_accuracy": 0.06819208525121212,
|
|
"num_tokens": 167482.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 9.973779201507568,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.802,
|
|
"mean_token_accuracy": 0.08643188402056694,
|
|
"num_tokens": 176598.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.932288265228271,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7014,
|
|
"mean_token_accuracy": 0.08371126018464566,
|
|
"num_tokens": 185883.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.903890705108642,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.672,
|
|
"mean_token_accuracy": 0.0767141304910183,
|
|
"num_tokens": 194820.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.905889225006103,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5994,
|
|
"mean_token_accuracy": 0.07663930654525757,
|
|
"num_tokens": 203710.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.7015700340271,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.4883,
|
|
"mean_token_accuracy": 0.08735396713018417,
|
|
"num_tokens": 212642.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.712564849853516,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4829,
|
|
"mean_token_accuracy": 0.07560812048614025,
|
|
"num_tokens": 221411.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.618709659576416,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.276,
|
|
"mean_token_accuracy": 0.07886423245072365,
|
|
"num_tokens": 230454.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.445035266876221,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2553,
|
|
"mean_token_accuracy": 0.08224238455295563,
|
|
"num_tokens": 240272.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.343283653259277,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.1773,
|
|
"mean_token_accuracy": 0.07860668338835239,
|
|
"num_tokens": 249652.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.141777420043946,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.0586,
|
|
"mean_token_accuracy": 0.08432014659047127,
|
|
"num_tokens": 258359.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.028361320495605,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.965,
|
|
"mean_token_accuracy": 0.08845296204090118,
|
|
"num_tokens": 266863.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.872007942199707,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9485,
|
|
"mean_token_accuracy": 0.08544271066784859,
|
|
"num_tokens": 274694.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.746376705169677,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.998,
|
|
"mean_token_accuracy": 0.07424286939203739,
|
|
"num_tokens": 285038.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.652623462677003,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9434,
|
|
"mean_token_accuracy": 0.07819288298487663,
|
|
"num_tokens": 293451.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.498311519622803,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7897,
|
|
"mean_token_accuracy": 0.08914674445986748,
|
|
"num_tokens": 302184.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.397662544250489,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8596,
|
|
"mean_token_accuracy": 0.08310422226786614,
|
|
"num_tokens": 311069.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.409068965911866,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8454,
|
|
"mean_token_accuracy": 0.07755065970122814,
|
|
"num_tokens": 319841.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.261888885498047,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7995,
|
|
"mean_token_accuracy": 0.0760527990758419,
|
|
"num_tokens": 328621.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.25572681427002,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7913,
|
|
"mean_token_accuracy": 0.07687021419405937,
|
|
"num_tokens": 336469.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.273696994781494,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.8148,
|
|
"mean_token_accuracy": 0.07753114774823189,
|
|
"num_tokens": 344859.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.1952486038208,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.8017,
|
|
"mean_token_accuracy": 0.08184871897101402,
|
|
"num_tokens": 354536.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.130875205993652,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.782,
|
|
"mean_token_accuracy": 0.08327300399541855,
|
|
"num_tokens": 363592.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.071674633026124,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7545,
|
|
"mean_token_accuracy": 0.08299842327833176,
|
|
"num_tokens": 373278.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.152774953842163,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.8002,
|
|
"mean_token_accuracy": 0.08091530725359916,
|
|
"num_tokens": 382498.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.122348976135253,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.7393,
|
|
"mean_token_accuracy": 0.09220762960612774,
|
|
"num_tokens": 391273.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.141310453414917,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.6711,
|
|
"mean_token_accuracy": 0.07944022566080093,
|
|
"num_tokens": 400340.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.08761625289917,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.7203,
|
|
"mean_token_accuracy": 0.07989789322018623,
|
|
"num_tokens": 409162.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.068865013122558,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.6955,
|
|
"mean_token_accuracy": 0.08263300582766533,
|
|
"num_tokens": 417345.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.020773887634277,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.6885,
|
|
"mean_token_accuracy": 0.08577688783407211,
|
|
"num_tokens": 427488.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.00288724899292,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.6381,
|
|
"mean_token_accuracy": 0.08082681894302368,
|
|
"num_tokens": 435496.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.024074983596801,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.592,
|
|
"mean_token_accuracy": 0.08299721553921699,
|
|
"num_tokens": 444290.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.002217769622803,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.7047,
|
|
"mean_token_accuracy": 0.0830993577837944,
|
|
"num_tokens": 452704.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.036895895004273,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.6246,
|
|
"mean_token_accuracy": 0.08470720425248146,
|
|
"num_tokens": 461445.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.038204240798951,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.6401,
|
|
"mean_token_accuracy": 0.08966793864965439,
|
|
"num_tokens": 470644.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.892484998703003,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.5766,
|
|
"mean_token_accuracy": 0.08981397226452828,
|
|
"num_tokens": 479212.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.9553313732147215,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.6849,
|
|
"mean_token_accuracy": 0.07991519421339036,
|
|
"num_tokens": 489027.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.001228666305542,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.7078,
|
|
"mean_token_accuracy": 0.0864037737250328,
|
|
"num_tokens": 497548.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.95409255027771,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.523,
|
|
"mean_token_accuracy": 0.09064189270138741,
|
|
"num_tokens": 505836.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.008459424972534,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.6857,
|
|
"mean_token_accuracy": 0.08223827481269837,
|
|
"num_tokens": 515463.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.042573165893554,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.5642,
|
|
"mean_token_accuracy": 0.08510619327425957,
|
|
"num_tokens": 524882.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.878521060943603,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.5627,
|
|
"mean_token_accuracy": 0.08412178531289101,
|
|
"num_tokens": 534523.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.9167151927948,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.5608,
|
|
"mean_token_accuracy": 0.08494351357221604,
|
|
"num_tokens": 543772.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.85640025138855,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.5374,
|
|
"mean_token_accuracy": 0.0830413281917572,
|
|
"num_tokens": 552024.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.888205242156983,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.5269,
|
|
"mean_token_accuracy": 0.08928481489419937,
|
|
"num_tokens": 561114.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.020092916488647,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.5728,
|
|
"mean_token_accuracy": 0.08862031325697899,
|
|
"num_tokens": 569655.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.867624950408936,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.5604,
|
|
"mean_token_accuracy": 0.08833226263523102,
|
|
"num_tokens": 578701.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.848216819763183,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.5257,
|
|
"mean_token_accuracy": 0.08904304206371308,
|
|
"num_tokens": 588166.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.775031900405883,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.5406,
|
|
"mean_token_accuracy": 0.08560147807002068,
|
|
"num_tokens": 598862.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.946314001083374,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.5054,
|
|
"mean_token_accuracy": 0.08752046972513199,
|
|
"num_tokens": 607680.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.798824310302734,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.5409,
|
|
"mean_token_accuracy": 0.08527569100260735,
|
|
"num_tokens": 617194.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.840095901489258,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.4681,
|
|
"mean_token_accuracy": 0.09199495688080787,
|
|
"num_tokens": 625946.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.863882684707642,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.4424,
|
|
"mean_token_accuracy": 0.09303685426712036,
|
|
"num_tokens": 634293.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.814202308654785,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.4365,
|
|
"mean_token_accuracy": 0.08837181255221367,
|
|
"num_tokens": 643479.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.780666065216065,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.4565,
|
|
"mean_token_accuracy": 0.0916847325861454,
|
|
"num_tokens": 653599.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.727724599838257,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.4396,
|
|
"mean_token_accuracy": 0.0935065358877182,
|
|
"num_tokens": 663042.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.7599810600280765,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.3799,
|
|
"mean_token_accuracy": 0.09194350391626357,
|
|
"num_tokens": 672254.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.850320053100586,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.4208,
|
|
"mean_token_accuracy": 0.09203241541981697,
|
|
"num_tokens": 681648.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.66558690071106,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.3756,
|
|
"mean_token_accuracy": 0.09696918800473213,
|
|
"num_tokens": 691080.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.681433391571045,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.4656,
|
|
"mean_token_accuracy": 0.08918402977287769,
|
|
"num_tokens": 700549.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.880253696441651,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5014,
|
|
"mean_token_accuracy": 0.09039791896939278,
|
|
"num_tokens": 710623.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.726108407974243,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.4269,
|
|
"mean_token_accuracy": 0.09598973840475082,
|
|
"num_tokens": 719695.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.658704042434692,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.3817,
|
|
"mean_token_accuracy": 0.09576035216450692,
|
|
"num_tokens": 728998.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.817303800582886,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.4597,
|
|
"mean_token_accuracy": 0.08954662233591079,
|
|
"num_tokens": 738537.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.673436260223388,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.4039,
|
|
"mean_token_accuracy": 0.092615757137537,
|
|
"num_tokens": 747555.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.643343782424926,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.3747,
|
|
"mean_token_accuracy": 0.09669034704566001,
|
|
"num_tokens": 756857.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.74475588798523,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.3802,
|
|
"mean_token_accuracy": 0.10021467804908753,
|
|
"num_tokens": 766181.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.649684762954712,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.4102,
|
|
"mean_token_accuracy": 0.09460910633206368,
|
|
"num_tokens": 774894.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.730149173736573,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.4062,
|
|
"mean_token_accuracy": 0.090732641518116,
|
|
"num_tokens": 784601.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.6217389583587645,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.4417,
|
|
"mean_token_accuracy": 0.09168312624096871,
|
|
"num_tokens": 794373.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.600076055526733,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.3455,
|
|
"mean_token_accuracy": 0.09434943050146102,
|
|
"num_tokens": 803608.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.752554702758789,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.3726,
|
|
"mean_token_accuracy": 0.09299151077866555,
|
|
"num_tokens": 812287.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.518798732757569,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.2985,
|
|
"mean_token_accuracy": 0.0987668864428997,
|
|
"num_tokens": 821049.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.701573610305786,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.3033,
|
|
"mean_token_accuracy": 0.10087452456355095,
|
|
"num_tokens": 829148.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.614718294143676,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.338,
|
|
"mean_token_accuracy": 0.09538405686616898,
|
|
"num_tokens": 838228.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.615123701095581,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.3246,
|
|
"mean_token_accuracy": 0.09597658589482308,
|
|
"num_tokens": 846059.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.705140447616577,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.4473,
|
|
"mean_token_accuracy": 0.09011114612221718,
|
|
"num_tokens": 856309.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.4943718910217285,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.2909,
|
|
"mean_token_accuracy": 0.1010310634970665,
|
|
"num_tokens": 865921.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.549269151687622,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.2654,
|
|
"mean_token_accuracy": 0.10411950796842576,
|
|
"num_tokens": 875076.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.629985094070435,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.3165,
|
|
"mean_token_accuracy": 0.0936420701444149,
|
|
"num_tokens": 884586.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.624928522109985,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.4262,
|
|
"mean_token_accuracy": 0.09749620333313942,
|
|
"num_tokens": 894192.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.579270219802856,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3251,
|
|
"mean_token_accuracy": 0.09591478407382965,
|
|
"num_tokens": 903542.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.532130336761474,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.259,
|
|
"mean_token_accuracy": 0.10112758129835128,
|
|
"num_tokens": 912625.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.499067693143277,
|
|
"eval_loss": 7.302952289581299,
|
|
"eval_mean_token_accuracy": 0.1013092825793334,
|
|
"eval_num_tokens": 912625.0,
|
|
"eval_runtime": 2.6557,
|
|
"eval_samples_per_second": 1336.368,
|
|
"eval_steps_per_second": 167.187,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.4851487159729,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.218,
|
|
"mean_token_accuracy": 0.10128417685627937,
|
|
"num_tokens": 921438.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.5242674350738525,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.2178,
|
|
"mean_token_accuracy": 0.10356576293706894,
|
|
"num_tokens": 931097.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.51934700012207,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.1837,
|
|
"mean_token_accuracy": 0.10203342065215111,
|
|
"num_tokens": 939375.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.429151105880737,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.1856,
|
|
"mean_token_accuracy": 0.10627232864499092,
|
|
"num_tokens": 948223.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.493127393722534,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.2191,
|
|
"mean_token_accuracy": 0.09962140917778015,
|
|
"num_tokens": 956243.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.528323316574097,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.2164,
|
|
"mean_token_accuracy": 0.10636536180973052,
|
|
"num_tokens": 964506.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.4207946300506595,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.2399,
|
|
"mean_token_accuracy": 0.0981321543455124,
|
|
"num_tokens": 973617.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.458613157272339,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.153,
|
|
"mean_token_accuracy": 0.10467413961887359,
|
|
"num_tokens": 981796.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.461965847015381,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.2143,
|
|
"mean_token_accuracy": 0.10574938952922822,
|
|
"num_tokens": 990677.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.328928756713867,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.1479,
|
|
"mean_token_accuracy": 0.10492404326796531,
|
|
"num_tokens": 1001131.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.405250549316406,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.2806,
|
|
"mean_token_accuracy": 0.0978497438132763,
|
|
"num_tokens": 1009925.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.450163650512695,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.211,
|
|
"mean_token_accuracy": 0.10683610290288925,
|
|
"num_tokens": 1018998.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.455226850509644,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.1873,
|
|
"mean_token_accuracy": 0.10439918413758278,
|
|
"num_tokens": 1027496.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.419484996795655,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.1247,
|
|
"mean_token_accuracy": 0.11058635860681534,
|
|
"num_tokens": 1035669.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.423511219024658,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.2032,
|
|
"mean_token_accuracy": 0.10203403383493423,
|
|
"num_tokens": 1044416.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.437924432754516,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.2185,
|
|
"mean_token_accuracy": 0.10163830146193505,
|
|
"num_tokens": 1053500.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.371481227874756,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.1261,
|
|
"mean_token_accuracy": 0.10348242074251175,
|
|
"num_tokens": 1061758.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.363918590545654,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.1857,
|
|
"mean_token_accuracy": 0.10252994000911712,
|
|
"num_tokens": 1070570.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.363972854614258,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.1311,
|
|
"mean_token_accuracy": 0.1051255501806736,
|
|
"num_tokens": 1080640.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.389889287948608,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.1887,
|
|
"mean_token_accuracy": 0.10394371747970581,
|
|
"num_tokens": 1090680.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.449554491043091,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.2423,
|
|
"mean_token_accuracy": 0.0991745762526989,
|
|
"num_tokens": 1100453.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.254623937606811,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.0726,
|
|
"mean_token_accuracy": 0.11043546348810196,
|
|
"num_tokens": 1109085.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.319861698150635,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.0439,
|
|
"mean_token_accuracy": 0.1101709671318531,
|
|
"num_tokens": 1118333.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.288142681121826,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.0361,
|
|
"mean_token_accuracy": 0.11464432403445243,
|
|
"num_tokens": 1127459.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.3332359790802,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.1691,
|
|
"mean_token_accuracy": 0.1035873532295227,
|
|
"num_tokens": 1137071.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.295236444473266,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.1535,
|
|
"mean_token_accuracy": 0.10194391012191772,
|
|
"num_tokens": 1146856.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.367660856246948,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.1897,
|
|
"mean_token_accuracy": 0.10307556018233299,
|
|
"num_tokens": 1155847.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.313199043273926,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.1504,
|
|
"mean_token_accuracy": 0.10430814623832703,
|
|
"num_tokens": 1164922.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.366647577285766,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.1199,
|
|
"mean_token_accuracy": 0.10304575935006141,
|
|
"num_tokens": 1174389.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.232296657562256,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.1016,
|
|
"mean_token_accuracy": 0.10503174364566803,
|
|
"num_tokens": 1182959.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.329212665557861,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.1193,
|
|
"mean_token_accuracy": 0.10655850470066071,
|
|
"num_tokens": 1191777.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.235855960845948,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.0269,
|
|
"mean_token_accuracy": 0.11145639792084694,
|
|
"num_tokens": 1200658.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.195139932632446,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.11196710541844368,
|
|
"num_tokens": 1209607.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.403189420700073,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.158,
|
|
"mean_token_accuracy": 0.10276976600289345,
|
|
"num_tokens": 1218146.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.122986936569214,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.0265,
|
|
"mean_token_accuracy": 0.11236484497785568,
|
|
"num_tokens": 1226886.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.288639545440674,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.0344,
|
|
"mean_token_accuracy": 0.11365340650081635,
|
|
"num_tokens": 1235408.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.159159898757935,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.025,
|
|
"mean_token_accuracy": 0.11552022844552993,
|
|
"num_tokens": 1244643.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.252866315841675,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.0859,
|
|
"mean_token_accuracy": 0.10846259817481041,
|
|
"num_tokens": 1254145.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.239425182342529,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.1242,
|
|
"mean_token_accuracy": 0.10573266297578812,
|
|
"num_tokens": 1263395.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.336990690231323,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.0478,
|
|
"mean_token_accuracy": 0.11208071708679199,
|
|
"num_tokens": 1272942.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.102862453460693,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.9782,
|
|
"mean_token_accuracy": 0.11494042649865151,
|
|
"num_tokens": 1281481.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.3168416023254395,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.1398,
|
|
"mean_token_accuracy": 0.10162141174077988,
|
|
"num_tokens": 1290455.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.213319635391235,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.0151,
|
|
"mean_token_accuracy": 0.1105075441300869,
|
|
"num_tokens": 1299241.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.2681577682495115,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.0957,
|
|
"mean_token_accuracy": 0.10155534744262695,
|
|
"num_tokens": 1308658.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.2813207626342775,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.0292,
|
|
"mean_token_accuracy": 0.10810133516788482,
|
|
"num_tokens": 1316987.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.062563180923462,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.984,
|
|
"mean_token_accuracy": 0.10933754518628121,
|
|
"num_tokens": 1325500.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.164379596710205,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.1053,
|
|
"mean_token_accuracy": 0.11041567921638488,
|
|
"num_tokens": 1335048.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.225457286834716,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.0434,
|
|
"mean_token_accuracy": 0.11181584149599075,
|
|
"num_tokens": 1343879.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.160865116119385,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9566,
|
|
"mean_token_accuracy": 0.11149153709411622,
|
|
"num_tokens": 1353169.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.17898383140564,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.0493,
|
|
"mean_token_accuracy": 0.10884664580225945,
|
|
"num_tokens": 1362912.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.154080009460449,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.053,
|
|
"mean_token_accuracy": 0.1088360756635666,
|
|
"num_tokens": 1372331.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.138390731811524,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.9359,
|
|
"mean_token_accuracy": 0.11739541292190551,
|
|
"num_tokens": 1381060.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.090542840957641,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.9446,
|
|
"mean_token_accuracy": 0.12408584356307983,
|
|
"num_tokens": 1388749.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.075214385986328,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.9269,
|
|
"mean_token_accuracy": 0.11776903718709945,
|
|
"num_tokens": 1398456.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.124542617797852,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.9221,
|
|
"mean_token_accuracy": 0.1167554147541523,
|
|
"num_tokens": 1407430.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.009809112548828,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.0097,
|
|
"mean_token_accuracy": 0.10950447097420693,
|
|
"num_tokens": 1416631.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.185323333740234,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9554,
|
|
"mean_token_accuracy": 0.1113646723330021,
|
|
"num_tokens": 1426124.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.108248376846314,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.0781,
|
|
"mean_token_accuracy": 0.10988424271345139,
|
|
"num_tokens": 1436510.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.196552371978759,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.007,
|
|
"mean_token_accuracy": 0.10850347205996513,
|
|
"num_tokens": 1445954.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.0683338165283205,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.0326,
|
|
"mean_token_accuracy": 0.10642682984471322,
|
|
"num_tokens": 1454981.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.130348634719849,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.981,
|
|
"mean_token_accuracy": 0.11366944611072541,
|
|
"num_tokens": 1464904.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.058192777633667,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.1203661397099495,
|
|
"num_tokens": 1473605.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.020400810241699,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8966,
|
|
"mean_token_accuracy": 0.12048028632998467,
|
|
"num_tokens": 1482989.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.021880626678467,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9765,
|
|
"mean_token_accuracy": 0.11234921589493752,
|
|
"num_tokens": 1492514.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.082749080657959,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.9776,
|
|
"mean_token_accuracy": 0.11630082428455353,
|
|
"num_tokens": 1501127.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.084704065322876,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.9896,
|
|
"mean_token_accuracy": 0.11336685940623284,
|
|
"num_tokens": 1509698.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.083193683624268,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.9501,
|
|
"mean_token_accuracy": 0.1161960445344448,
|
|
"num_tokens": 1518224.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.068277740478516,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.9509,
|
|
"mean_token_accuracy": 0.11460287198424339,
|
|
"num_tokens": 1527163.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.035658502578736,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.9773,
|
|
"mean_token_accuracy": 0.11205969154834747,
|
|
"num_tokens": 1536980.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.014459228515625,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.933,
|
|
"mean_token_accuracy": 0.1107229508459568,
|
|
"num_tokens": 1546388.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.04270453453064,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.9719,
|
|
"mean_token_accuracy": 0.1091302290558815,
|
|
"num_tokens": 1555618.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.163297176361084,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.0471,
|
|
"mean_token_accuracy": 0.11037192791700363,
|
|
"num_tokens": 1565398.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.015673398971558,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9003,
|
|
"mean_token_accuracy": 0.11521378830075264,
|
|
"num_tokens": 1574052.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.948957300186157,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8385,
|
|
"mean_token_accuracy": 0.12051832377910614,
|
|
"num_tokens": 1582930.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.9927011013031,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.9591,
|
|
"mean_token_accuracy": 0.11156816780567169,
|
|
"num_tokens": 1592007.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.008735847473145,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.7845,
|
|
"mean_token_accuracy": 0.12361353784799575,
|
|
"num_tokens": 1600532.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.962292051315307,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9289,
|
|
"mean_token_accuracy": 0.1097280241549015,
|
|
"num_tokens": 1610319.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.939612913131714,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.7976,
|
|
"mean_token_accuracy": 0.11956171989440918,
|
|
"num_tokens": 1618161.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.0276683330535885,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.9023,
|
|
"mean_token_accuracy": 0.11797274127602578,
|
|
"num_tokens": 1627676.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.978840589523315,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.8601,
|
|
"mean_token_accuracy": 0.1139043152332306,
|
|
"num_tokens": 1636693.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.861077356338501,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.8644,
|
|
"mean_token_accuracy": 0.11552541553974152,
|
|
"num_tokens": 1645577.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.998029899597168,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8572,
|
|
"mean_token_accuracy": 0.1161652997136116,
|
|
"num_tokens": 1654239.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.036835432052612,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.8756,
|
|
"mean_token_accuracy": 0.11719652488827706,
|
|
"num_tokens": 1663085.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.875935173034668,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.9058,
|
|
"mean_token_accuracy": 0.11936888322234154,
|
|
"num_tokens": 1671621.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.0160260677337645,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.8876,
|
|
"mean_token_accuracy": 0.11704642698168755,
|
|
"num_tokens": 1680912.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.802255773544312,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.7984,
|
|
"mean_token_accuracy": 0.1263578489422798,
|
|
"num_tokens": 1689772.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.00699782371521,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.9767,
|
|
"mean_token_accuracy": 0.10714940577745438,
|
|
"num_tokens": 1698985.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.9496043682098385,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.8334,
|
|
"mean_token_accuracy": 0.11713856160640716,
|
|
"num_tokens": 1707365.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.864590072631836,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.8486,
|
|
"mean_token_accuracy": 0.12177055031061172,
|
|
"num_tokens": 1715858.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.043800115585327,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.8787,
|
|
"mean_token_accuracy": 0.11329701766371728,
|
|
"num_tokens": 1724721.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.8629217624664305,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.8973,
|
|
"mean_token_accuracy": 0.11501588821411132,
|
|
"num_tokens": 1733700.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.911140823364258,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8497,
|
|
"mean_token_accuracy": 0.11782671734690667,
|
|
"num_tokens": 1742167.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.022890186309814,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8634,
|
|
"mean_token_accuracy": 0.11890430226922036,
|
|
"num_tokens": 1752087.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.841674852371216,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.8119,
|
|
"mean_token_accuracy": 0.12273848205804824,
|
|
"num_tokens": 1760749.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.708367443084716,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.8394,
|
|
"mean_token_accuracy": 0.118587414175272,
|
|
"num_tokens": 1770923.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.947722911834717,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.8257,
|
|
"mean_token_accuracy": 0.11907628253102302,
|
|
"num_tokens": 1779233.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.948696708679199,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.8234,
|
|
"mean_token_accuracy": 0.11929655894637108,
|
|
"num_tokens": 1787593.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.848589468002319,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8683,
|
|
"mean_token_accuracy": 0.11564295887947082,
|
|
"num_tokens": 1796655.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.848339986801148,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.7993,
|
|
"mean_token_accuracy": 0.1252160184085369,
|
|
"num_tokens": 1805790.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.861878204345703,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.7821,
|
|
"mean_token_accuracy": 0.11595688387751579,
|
|
"num_tokens": 1813776.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.850721193863465,
|
|
"eval_loss": 6.848105430603027,
|
|
"eval_mean_token_accuracy": 0.12149045062628952,
|
|
"eval_num_tokens": 1813776.0,
|
|
"eval_runtime": 2.8306,
|
|
"eval_samples_per_second": 1253.81,
|
|
"eval_steps_per_second": 156.859,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.787899780273437,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.7924,
|
|
"mean_token_accuracy": 0.11906375139951705,
|
|
"num_tokens": 1823034.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.925178575515747,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.8714,
|
|
"mean_token_accuracy": 0.115791717171669,
|
|
"num_tokens": 1832320.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.781938219070435,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.7573,
|
|
"mean_token_accuracy": 0.12310137376189231,
|
|
"num_tokens": 1841165.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.863918399810791,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.8264,
|
|
"mean_token_accuracy": 0.11974046677350998,
|
|
"num_tokens": 1850440.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.80152702331543,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.8336,
|
|
"mean_token_accuracy": 0.11709848865866661,
|
|
"num_tokens": 1859304.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.884733772277832,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.7027,
|
|
"mean_token_accuracy": 0.12448290213942528,
|
|
"num_tokens": 1867560.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.801468753814698,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.7809,
|
|
"mean_token_accuracy": 0.12414068430662155,
|
|
"num_tokens": 1876742.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.889640283584595,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.7848,
|
|
"mean_token_accuracy": 0.12225122228264809,
|
|
"num_tokens": 1885738.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.826289463043213,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.8167,
|
|
"mean_token_accuracy": 0.11297583729028701,
|
|
"num_tokens": 1894688.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.887860727310181,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.8767,
|
|
"mean_token_accuracy": 0.12184796631336212,
|
|
"num_tokens": 1903809.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.873423671722412,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.834,
|
|
"mean_token_accuracy": 0.1166172593832016,
|
|
"num_tokens": 1913558.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.811271381378174,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.8109,
|
|
"mean_token_accuracy": 0.11310545653104782,
|
|
"num_tokens": 1923324.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.750212335586548,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.7839,
|
|
"mean_token_accuracy": 0.120677300542593,
|
|
"num_tokens": 1932849.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.9223932266235355,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.9007,
|
|
"mean_token_accuracy": 0.11593888625502587,
|
|
"num_tokens": 1942678.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.778281164169312,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.7536,
|
|
"mean_token_accuracy": 0.11781821995973588,
|
|
"num_tokens": 1951670.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.896324634552002,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.8645,
|
|
"mean_token_accuracy": 0.12120396941900254,
|
|
"num_tokens": 1961060.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.905940532684326,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.8059,
|
|
"mean_token_accuracy": 0.11919425427913666,
|
|
"num_tokens": 1970790.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.860645055770874,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.8232,
|
|
"mean_token_accuracy": 0.11798580288887024,
|
|
"num_tokens": 1980016.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.765835952758789,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.7942,
|
|
"mean_token_accuracy": 0.12255456075072288,
|
|
"num_tokens": 1989542.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.922926950454712,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.7695,
|
|
"mean_token_accuracy": 0.1269351787865162,
|
|
"num_tokens": 1999212.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.666564416885376,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.7551,
|
|
"mean_token_accuracy": 0.12631533294916153,
|
|
"num_tokens": 2008378.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.820988655090332,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.8346,
|
|
"mean_token_accuracy": 0.11563185527920723,
|
|
"num_tokens": 2018985.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.789529275894165,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.7218,
|
|
"mean_token_accuracy": 0.12006904929876328,
|
|
"num_tokens": 2027274.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.760359621047973,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.8206,
|
|
"mean_token_accuracy": 0.12053585797548294,
|
|
"num_tokens": 2036892.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.926661109924316,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.7966,
|
|
"mean_token_accuracy": 0.11386888101696968,
|
|
"num_tokens": 2046042.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.7287068367004395,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.7663,
|
|
"mean_token_accuracy": 0.11801247894763947,
|
|
"num_tokens": 2054576.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.731119585037232,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.7113,
|
|
"mean_token_accuracy": 0.12452540397644044,
|
|
"num_tokens": 2063552.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.721849632263184,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.7168,
|
|
"mean_token_accuracy": 0.12163661047816277,
|
|
"num_tokens": 2072965.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.808212089538574,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.7754,
|
|
"mean_token_accuracy": 0.12033762335777283,
|
|
"num_tokens": 2082461.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.747566175460816,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.7715,
|
|
"mean_token_accuracy": 0.12163460701704025,
|
|
"num_tokens": 2091022.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.744241046905517,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.6552,
|
|
"mean_token_accuracy": 0.13288048580288886,
|
|
"num_tokens": 2099896.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.769083023071289,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.7577,
|
|
"mean_token_accuracy": 0.12401864305138588,
|
|
"num_tokens": 2109122.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.859591378106011,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.7057,
|
|
"mean_token_accuracy": 0.12527080294158724,
|
|
"num_tokens": 2117879.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.636478424072266,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.4751,
|
|
"mean_token_accuracy": 0.13114811182022096,
|
|
"num_tokens": 2127552.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.738952016830444,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.4838,
|
|
"mean_token_accuracy": 0.12866360768675805,
|
|
"num_tokens": 2136648.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.67201509475708,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.5263,
|
|
"mean_token_accuracy": 0.13279551938176154,
|
|
"num_tokens": 2144852.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.675800752639771,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.5531,
|
|
"mean_token_accuracy": 0.12884859815239907,
|
|
"num_tokens": 2154567.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.688924407958984,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.5359,
|
|
"mean_token_accuracy": 0.12873365879058837,
|
|
"num_tokens": 2164260.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.644716072082519,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.5181,
|
|
"mean_token_accuracy": 0.12822481170296668,
|
|
"num_tokens": 2173247.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.574692296981811,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.5603,
|
|
"mean_token_accuracy": 0.12655097916722297,
|
|
"num_tokens": 2181595.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.753303909301758,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.4769,
|
|
"mean_token_accuracy": 0.12973184511065483,
|
|
"num_tokens": 2190608.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.538211107254028,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.4797,
|
|
"mean_token_accuracy": 0.14387280195951463,
|
|
"num_tokens": 2199131.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.674481058120728,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.6247,
|
|
"mean_token_accuracy": 0.12701441198587418,
|
|
"num_tokens": 2208365.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.658597755432129,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.4953,
|
|
"mean_token_accuracy": 0.1288551889359951,
|
|
"num_tokens": 2216851.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.630938100814819,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.5736,
|
|
"mean_token_accuracy": 0.12576404958963394,
|
|
"num_tokens": 2225309.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.695171594619751,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.5228,
|
|
"mean_token_accuracy": 0.1320165403187275,
|
|
"num_tokens": 2235036.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.625878429412841,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.5198,
|
|
"mean_token_accuracy": 0.12893396690487863,
|
|
"num_tokens": 2244038.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.645451402664184,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.4709,
|
|
"mean_token_accuracy": 0.13499100208282472,
|
|
"num_tokens": 2252828.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.610739994049072,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.13818272650241853,
|
|
"num_tokens": 2261709.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.59524974822998,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.4923,
|
|
"mean_token_accuracy": 0.13403210416436195,
|
|
"num_tokens": 2269801.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.591278791427612,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.4461,
|
|
"mean_token_accuracy": 0.1370372988283634,
|
|
"num_tokens": 2280587.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.460198545455933,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.1323759824037552,
|
|
"num_tokens": 2289783.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.588274049758911,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.5313,
|
|
"mean_token_accuracy": 0.12670424431562424,
|
|
"num_tokens": 2298785.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.64993691444397,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.6208,
|
|
"mean_token_accuracy": 0.13185966834425927,
|
|
"num_tokens": 2308400.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.641791296005249,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.4313,
|
|
"mean_token_accuracy": 0.13738049939274788,
|
|
"num_tokens": 2317440.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.503924989700318,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.4783,
|
|
"mean_token_accuracy": 0.131829671561718,
|
|
"num_tokens": 2326730.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.57679181098938,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.13649967312812805,
|
|
"num_tokens": 2336126.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.6197991371154785,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.4954,
|
|
"mean_token_accuracy": 0.13550228029489517,
|
|
"num_tokens": 2345144.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.5108154773712155,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.444,
|
|
"mean_token_accuracy": 0.13719593361020088,
|
|
"num_tokens": 2353851.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.5021027565002445,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.4387,
|
|
"mean_token_accuracy": 0.13497617840766907,
|
|
"num_tokens": 2362316.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.660447835922241,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.5457,
|
|
"mean_token_accuracy": 0.1298186369240284,
|
|
"num_tokens": 2373350.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.587765121459961,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.4847,
|
|
"mean_token_accuracy": 0.13602642267942427,
|
|
"num_tokens": 2381753.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.5865904808044435,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.5174,
|
|
"mean_token_accuracy": 0.1304781563580036,
|
|
"num_tokens": 2390093.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.707785511016846,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.5307,
|
|
"mean_token_accuracy": 0.13693483620882035,
|
|
"num_tokens": 2399103.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.539694452285767,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.4963,
|
|
"mean_token_accuracy": 0.13442791253328323,
|
|
"num_tokens": 2407530.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.5834043502807615,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.4721,
|
|
"mean_token_accuracy": 0.13653133809566498,
|
|
"num_tokens": 2416264.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.560483932495117,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.5643,
|
|
"mean_token_accuracy": 0.1255203329026699,
|
|
"num_tokens": 2425256.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.543246126174926,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.5303,
|
|
"mean_token_accuracy": 0.12744321376085282,
|
|
"num_tokens": 2434343.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.625397109985352,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.439,
|
|
"mean_token_accuracy": 0.13895059525966644,
|
|
"num_tokens": 2442864.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.505967378616333,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.482,
|
|
"mean_token_accuracy": 0.1311720348894596,
|
|
"num_tokens": 2451701.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.503494215011597,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.4696,
|
|
"mean_token_accuracy": 0.13065513744950294,
|
|
"num_tokens": 2460943.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.595712518692016,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.4954,
|
|
"mean_token_accuracy": 0.13297019228339196,
|
|
"num_tokens": 2470294.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.517033290863037,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.13079115897417068,
|
|
"num_tokens": 2479530.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.589979982376098,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.3268,
|
|
"mean_token_accuracy": 0.13975912407040597,
|
|
"num_tokens": 2488511.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.550735712051392,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.5284,
|
|
"mean_token_accuracy": 0.1279216580092907,
|
|
"num_tokens": 2497868.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.562166547775268,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.4477,
|
|
"mean_token_accuracy": 0.13719073608517646,
|
|
"num_tokens": 2506873.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.647982025146485,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.5799,
|
|
"mean_token_accuracy": 0.13164269253611566,
|
|
"num_tokens": 2517317.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.559144067764282,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.4983,
|
|
"mean_token_accuracy": 0.13328756764531136,
|
|
"num_tokens": 2527348.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.5675629615783695,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.4813,
|
|
"mean_token_accuracy": 0.13246413692831993,
|
|
"num_tokens": 2536277.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.5679216384887695,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.5581,
|
|
"mean_token_accuracy": 0.1256929397583008,
|
|
"num_tokens": 2545638.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.445409727096558,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.3901,
|
|
"mean_token_accuracy": 0.13980056941509247,
|
|
"num_tokens": 2555694.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.59971604347229,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.4436,
|
|
"mean_token_accuracy": 0.13092342391610146,
|
|
"num_tokens": 2564344.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.468144750595092,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.4329,
|
|
"mean_token_accuracy": 0.13643699064850806,
|
|
"num_tokens": 2573726.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.531007194519043,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.13096859753131868,
|
|
"num_tokens": 2582845.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.531826734542847,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.3872,
|
|
"mean_token_accuracy": 0.1383131504058838,
|
|
"num_tokens": 2591258.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.575545120239258,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.4782,
|
|
"mean_token_accuracy": 0.13248563408851624,
|
|
"num_tokens": 2600099.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.531699705123901,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.4466,
|
|
"mean_token_accuracy": 0.1367930218577385,
|
|
"num_tokens": 2608845.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.4251275062561035,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.3915,
|
|
"mean_token_accuracy": 0.14260655716061593,
|
|
"num_tokens": 2618723.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.515915298461914,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.4512,
|
|
"mean_token_accuracy": 0.13224708959460257,
|
|
"num_tokens": 2628542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.4922788619995115,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.4214,
|
|
"mean_token_accuracy": 0.14079917669296266,
|
|
"num_tokens": 2637395.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.43779935836792,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.2848,
|
|
"mean_token_accuracy": 0.14780827164649962,
|
|
"num_tokens": 2645545.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.449300241470337,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.4446,
|
|
"mean_token_accuracy": 0.1389339432120323,
|
|
"num_tokens": 2654000.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.54451584815979,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.4506,
|
|
"mean_token_accuracy": 0.13394884243607522,
|
|
"num_tokens": 2664017.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.453659868240356,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.4412,
|
|
"mean_token_accuracy": 0.13818295150995255,
|
|
"num_tokens": 2673250.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.53358006477356,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.4796,
|
|
"mean_token_accuracy": 0.13405440375208855,
|
|
"num_tokens": 2682190.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.550917911529541,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.4152,
|
|
"mean_token_accuracy": 0.13004215285182,
|
|
"num_tokens": 2691425.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.5540321350097654,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.5022,
|
|
"mean_token_accuracy": 0.1328156165778637,
|
|
"num_tokens": 2701123.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.401340389251709,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.5081,
|
|
"mean_token_accuracy": 0.1309308812022209,
|
|
"num_tokens": 2710829.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.600834655761719,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.4494,
|
|
"mean_token_accuracy": 0.13603252917528152,
|
|
"num_tokens": 2720313.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.446039533615112,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.3644,
|
|
"mean_token_accuracy": 0.13654347658157348,
|
|
"num_tokens": 2729734.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.42241061270774,
|
|
"eval_loss": 6.562856674194336,
|
|
"eval_mean_token_accuracy": 0.13629841859880332,
|
|
"eval_num_tokens": 2729734.0,
|
|
"eval_runtime": 2.6412,
|
|
"eval_samples_per_second": 1343.717,
|
|
"eval_steps_per_second": 168.107,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.496413612365723,
|
|
"epoch": 1.2930812204555222,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497508598642197,
|
|
"loss": 6.4498,
|
|
"mean_token_accuracy": 0.13591181635856628,
|
|
"num_tokens": 2739001.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.4927387714385985,
|
|
"epoch": 1.2973785990545768,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497459014588216,
|
|
"loss": 6.3461,
|
|
"mean_token_accuracy": 0.14019913077354432,
|
|
"num_tokens": 2748460.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.442285966873169,
|
|
"epoch": 1.3016759776536313,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497408944772663,
|
|
"loss": 6.4495,
|
|
"mean_token_accuracy": 0.13356760740280152,
|
|
"num_tokens": 2757993.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.442544317245483,
|
|
"epoch": 1.3059733562526858,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004973583893048707,
|
|
"loss": 6.4679,
|
|
"mean_token_accuracy": 0.1391291558742523,
|
|
"num_tokens": 2767380.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.537827825546264,
|
|
"epoch": 1.3102707348517404,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004973073482952321,
|
|
"loss": 6.3971,
|
|
"mean_token_accuracy": 0.1322616696357727,
|
|
"num_tokens": 2776207.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.452885007858276,
|
|
"epoch": 1.314568113450795,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004972558218552004,
|
|
"loss": 6.4398,
|
|
"mean_token_accuracy": 0.13432919085025788,
|
|
"num_tokens": 2785857.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.486521863937378,
|
|
"epoch": 1.3188654920498495,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004972038100972885,
|
|
"loss": 6.5342,
|
|
"mean_token_accuracy": 0.1303657077252865,
|
|
"num_tokens": 2795500.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.549186372756958,
|
|
"epoch": 1.323162870648904,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004971513131350697,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.13469692692160606,
|
|
"num_tokens": 2804188.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.489724349975586,
|
|
"epoch": 1.3274602492479588,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004970983310831759,
|
|
"loss": 6.424,
|
|
"mean_token_accuracy": 0.12920740991830826,
|
|
"num_tokens": 2814255.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.468247175216675,
|
|
"epoch": 1.3317576278470133,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004970448640572989,
|
|
"loss": 6.3908,
|
|
"mean_token_accuracy": 0.13704784587025642,
|
|
"num_tokens": 2823059.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.443971824645996,
|
|
"epoch": 1.336055006446068,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004969909121741895,
|
|
"loss": 6.3212,
|
|
"mean_token_accuracy": 0.14167552590370178,
|
|
"num_tokens": 2831729.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.461103248596191,
|
|
"epoch": 1.3403523850451224,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004969364755516569,
|
|
"loss": 6.3786,
|
|
"mean_token_accuracy": 0.1437497019767761,
|
|
"num_tokens": 2839597.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.413368654251099,
|
|
"epoch": 1.344649763644177,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004968815543085689,
|
|
"loss": 6.2597,
|
|
"mean_token_accuracy": 0.1471225917339325,
|
|
"num_tokens": 2847964.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.442201375961304,
|
|
"epoch": 1.3489471422432318,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004968261485648516,
|
|
"loss": 6.4682,
|
|
"mean_token_accuracy": 0.13341710045933725,
|
|
"num_tokens": 2857712.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.399351978302002,
|
|
"epoch": 1.3532445208422863,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000496770258441489,
|
|
"loss": 6.4641,
|
|
"mean_token_accuracy": 0.13078686520457267,
|
|
"num_tokens": 2866724.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.548488187789917,
|
|
"epoch": 1.3575418994413408,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004967138840605228,
|
|
"loss": 6.3578,
|
|
"mean_token_accuracy": 0.14059459790587425,
|
|
"num_tokens": 2875318.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.420933103561401,
|
|
"epoch": 1.3618392780403954,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496657025545052,
|
|
"loss": 6.3964,
|
|
"mean_token_accuracy": 0.13322596848011017,
|
|
"num_tokens": 2883794.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.4831382751464846,
|
|
"epoch": 1.36613665663945,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000496599683019233,
|
|
"loss": 6.4175,
|
|
"mean_token_accuracy": 0.1317307434976101,
|
|
"num_tokens": 2891996.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.506607484817505,
|
|
"epoch": 1.3704340352385045,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000496541856608279,
|
|
"loss": 6.5069,
|
|
"mean_token_accuracy": 0.13520395755767822,
|
|
"num_tokens": 2901718.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.488222885131836,
|
|
"epoch": 1.374731413837559,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004964835464384595,
|
|
"loss": 6.3369,
|
|
"mean_token_accuracy": 0.1403390385210514,
|
|
"num_tokens": 2910924.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.356355381011963,
|
|
"epoch": 1.3790287924366136,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000496424752637101,
|
|
"loss": 6.4404,
|
|
"mean_token_accuracy": 0.1325266629457474,
|
|
"num_tokens": 2920075.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.4637175559997555,
|
|
"epoch": 1.3833261710356681,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004963654753325853,
|
|
"loss": 6.3958,
|
|
"mean_token_accuracy": 0.14068760275840758,
|
|
"num_tokens": 2928946.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.382746267318725,
|
|
"epoch": 1.387623549634723,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004963057146543505,
|
|
"loss": 6.3788,
|
|
"mean_token_accuracy": 0.14359928518533707,
|
|
"num_tokens": 2936817.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.462194156646729,
|
|
"epoch": 1.3919209282337774,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00049624547073289,
|
|
"loss": 6.3627,
|
|
"mean_token_accuracy": 0.1397081971168518,
|
|
"num_tokens": 2946096.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.3689014434814455,
|
|
"epoch": 1.396218306832832,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004961847436997526,
|
|
"loss": 6.2999,
|
|
"mean_token_accuracy": 0.147130286693573,
|
|
"num_tokens": 2955632.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.4761738777160645,
|
|
"epoch": 1.4005156854318865,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004961235336875416,
|
|
"loss": 6.3289,
|
|
"mean_token_accuracy": 0.14046488851308822,
|
|
"num_tokens": 2963779.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.419768238067627,
|
|
"epoch": 1.404813064030941,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004960618408299154,
|
|
"loss": 6.3199,
|
|
"mean_token_accuracy": 0.14862223267555236,
|
|
"num_tokens": 2972778.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.360955905914307,
|
|
"epoch": 1.4091104426299956,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004959996652615865,
|
|
"loss": 6.4341,
|
|
"mean_token_accuracy": 0.1319722667336464,
|
|
"num_tokens": 2982318.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.514253997802735,
|
|
"epoch": 1.4134078212290504,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959370071183216,
|
|
"loss": 6.4039,
|
|
"mean_token_accuracy": 0.14093454405665398,
|
|
"num_tokens": 2991822.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.470624685287476,
|
|
"epoch": 1.417705199828105,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004958738665369407,
|
|
"loss": 6.4473,
|
|
"mean_token_accuracy": 0.13844559118151664,
|
|
"num_tokens": 3001203.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.412876272201538,
|
|
"epoch": 1.4220025784271595,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004958102436553179,
|
|
"loss": 6.3646,
|
|
"mean_token_accuracy": 0.13862057998776436,
|
|
"num_tokens": 3010445.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.440686845779419,
|
|
"epoch": 1.426299957026214,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00049574613861238,
|
|
"loss": 6.3898,
|
|
"mean_token_accuracy": 0.13541134893894197,
|
|
"num_tokens": 3019919.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.509662914276123,
|
|
"epoch": 1.4305973356252686,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004956815515481069,
|
|
"loss": 6.5165,
|
|
"mean_token_accuracy": 0.12789023146033288,
|
|
"num_tokens": 3029469.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.461795949935913,
|
|
"epoch": 1.4348947142243231,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004956164826035309,
|
|
"loss": 6.3809,
|
|
"mean_token_accuracy": 0.14309595823287963,
|
|
"num_tokens": 3038627.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.41576566696167,
|
|
"epoch": 1.4391920928233777,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004955509319207363,
|
|
"loss": 6.4312,
|
|
"mean_token_accuracy": 0.13531435057520866,
|
|
"num_tokens": 3048124.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.411845588684082,
|
|
"epoch": 1.4434894714224322,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004954848996428601,
|
|
"loss": 6.269,
|
|
"mean_token_accuracy": 0.14824977070093154,
|
|
"num_tokens": 3056394.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.383704996109008,
|
|
"epoch": 1.4477868500214868,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00049541838591409,
|
|
"loss": 6.3767,
|
|
"mean_token_accuracy": 0.1388603314757347,
|
|
"num_tokens": 3066521.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.452395057678222,
|
|
"epoch": 1.4520842286205415,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004953513908796657,
|
|
"loss": 6.4368,
|
|
"mean_token_accuracy": 0.13928449675440788,
|
|
"num_tokens": 3076571.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.397390127182007,
|
|
"epoch": 1.456381607219596,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004952839146858773,
|
|
"loss": 6.4782,
|
|
"mean_token_accuracy": 0.1314165897667408,
|
|
"num_tokens": 3086086.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.3878926753997805,
|
|
"epoch": 1.4606789858186506,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004952159574800658,
|
|
"loss": 6.3599,
|
|
"mean_token_accuracy": 0.14619865864515305,
|
|
"num_tokens": 3095046.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.403419876098633,
|
|
"epoch": 1.4649763644177052,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004951475194106229,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.13754306733608246,
|
|
"num_tokens": 3104193.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.475354766845703,
|
|
"epoch": 1.4692737430167597,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004950786006269898,
|
|
"loss": 6.4103,
|
|
"mean_token_accuracy": 0.1397714652121067,
|
|
"num_tokens": 3112965.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.420578479766846,
|
|
"epoch": 1.4735711216158143,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004950092012796576,
|
|
"loss": 6.4802,
|
|
"mean_token_accuracy": 0.13525724932551383,
|
|
"num_tokens": 3122883.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.446125411987305,
|
|
"epoch": 1.477868500214869,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004949393215201666,
|
|
"loss": 6.334,
|
|
"mean_token_accuracy": 0.14860173985362052,
|
|
"num_tokens": 3132182.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.438638639450073,
|
|
"epoch": 1.4821658788139236,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004948689615011065,
|
|
"loss": 6.3881,
|
|
"mean_token_accuracy": 0.13537878468632697,
|
|
"num_tokens": 3141957.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.339945554733276,
|
|
"epoch": 1.4864632574129781,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004947981213761154,
|
|
"loss": 6.3201,
|
|
"mean_token_accuracy": 0.1484260804951191,
|
|
"num_tokens": 3151374.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.324282693862915,
|
|
"epoch": 1.4907606360120327,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004947268012998797,
|
|
"loss": 6.3382,
|
|
"mean_token_accuracy": 0.13972473591566087,
|
|
"num_tokens": 3160815.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.4526674270629885,
|
|
"epoch": 1.4950580146110872,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000494655001428134,
|
|
"loss": 6.345,
|
|
"mean_token_accuracy": 0.14593861922621726,
|
|
"num_tokens": 3169684.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.443645286560058,
|
|
"epoch": 1.4993553932101418,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945827219176604,
|
|
"loss": 6.3344,
|
|
"mean_token_accuracy": 0.13984118551015853,
|
|
"num_tokens": 3178754.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.419276237487793,
|
|
"epoch": 1.5036527718091963,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004945099629262888,
|
|
"loss": 6.4109,
|
|
"mean_token_accuracy": 0.1383548989892006,
|
|
"num_tokens": 3187315.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.409943437576294,
|
|
"epoch": 1.5079501504082509,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004944367246128954,
|
|
"loss": 6.2452,
|
|
"mean_token_accuracy": 0.1493537425994873,
|
|
"num_tokens": 3195975.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.326399850845337,
|
|
"epoch": 1.5122475290073054,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004943630071374036,
|
|
"loss": 6.3539,
|
|
"mean_token_accuracy": 0.14070346504449843,
|
|
"num_tokens": 3204071.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.41026463508606,
|
|
"epoch": 1.51654490760636,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004942888106607828,
|
|
"loss": 6.339,
|
|
"mean_token_accuracy": 0.14475558772683145,
|
|
"num_tokens": 3213001.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.4736912727355955,
|
|
"epoch": 1.5208422862054147,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004942141353450486,
|
|
"loss": 6.3736,
|
|
"mean_token_accuracy": 0.13328379541635513,
|
|
"num_tokens": 3222576.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.408621740341187,
|
|
"epoch": 1.5251396648044693,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004941389813532619,
|
|
"loss": 6.3922,
|
|
"mean_token_accuracy": 0.1397964984178543,
|
|
"num_tokens": 3230474.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.462115287780762,
|
|
"epoch": 1.5294370434035238,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000494063348849529,
|
|
"loss": 6.3624,
|
|
"mean_token_accuracy": 0.14171589016914368,
|
|
"num_tokens": 3238723.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.310955953598023,
|
|
"epoch": 1.5337344220025786,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004939872379990011,
|
|
"loss": 6.3057,
|
|
"mean_token_accuracy": 0.1414056584239006,
|
|
"num_tokens": 3247844.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.499557447433472,
|
|
"epoch": 1.5380318006016331,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004939106489678739,
|
|
"loss": 6.3996,
|
|
"mean_token_accuracy": 0.13961976021528244,
|
|
"num_tokens": 3257228.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.428326511383057,
|
|
"epoch": 1.5423291792006877,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000493833581923387,
|
|
"loss": 6.3806,
|
|
"mean_token_accuracy": 0.1350629784166813,
|
|
"num_tokens": 3266655.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.40258674621582,
|
|
"epoch": 1.5466265577997422,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004937560370338244,
|
|
"loss": 6.3076,
|
|
"mean_token_accuracy": 0.13736466988921164,
|
|
"num_tokens": 3275213.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.4639317989349365,
|
|
"epoch": 1.5509239363987968,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000493678014468513,
|
|
"loss": 6.4331,
|
|
"mean_token_accuracy": 0.13448011130094528,
|
|
"num_tokens": 3284447.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.456356191635132,
|
|
"epoch": 1.5552213149978513,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004935995143978227,
|
|
"loss": 6.3597,
|
|
"mean_token_accuracy": 0.13500829488039018,
|
|
"num_tokens": 3293568.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.347956132888794,
|
|
"epoch": 1.5595186935969059,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004935205369931664,
|
|
"loss": 6.2955,
|
|
"mean_token_accuracy": 0.14433372840285302,
|
|
"num_tokens": 3302055.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.415659618377686,
|
|
"epoch": 1.5638160721959604,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004934410824269992,
|
|
"loss": 6.3521,
|
|
"mean_token_accuracy": 0.141788549721241,
|
|
"num_tokens": 3310672.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.385811614990234,
|
|
"epoch": 1.568113450795015,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004933611508728182,
|
|
"loss": 6.3045,
|
|
"mean_token_accuracy": 0.14355277717113496,
|
|
"num_tokens": 3319666.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.430783605575561,
|
|
"epoch": 1.5724108293940695,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000493280742505162,
|
|
"loss": 6.3168,
|
|
"mean_token_accuracy": 0.14499156102538108,
|
|
"num_tokens": 3329121.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.328087186813354,
|
|
"epoch": 1.576708207993124,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004931998574996102,
|
|
"loss": 6.2982,
|
|
"mean_token_accuracy": 0.14460441917181016,
|
|
"num_tokens": 3337868.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.422233009338379,
|
|
"epoch": 1.5810055865921788,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004931184960327832,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.14177661910653114,
|
|
"num_tokens": 3346447.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.337698698043823,
|
|
"epoch": 1.5853029651912334,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004930366582823421,
|
|
"loss": 6.2971,
|
|
"mean_token_accuracy": 0.13885951936244964,
|
|
"num_tokens": 3355187.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.434715986251831,
|
|
"epoch": 1.589600343790288,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004929543444269879,
|
|
"loss": 6.5041,
|
|
"mean_token_accuracy": 0.12198112457990647,
|
|
"num_tokens": 3365978.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.421086025238037,
|
|
"epoch": 1.5938977223893425,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492871554646461,
|
|
"loss": 6.4553,
|
|
"mean_token_accuracy": 0.13949850127100943,
|
|
"num_tokens": 3374607.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.452510786056519,
|
|
"epoch": 1.5981951009883972,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004927882891215413,
|
|
"loss": 6.3641,
|
|
"mean_token_accuracy": 0.14000257328152657,
|
|
"num_tokens": 3384541.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.383066320419312,
|
|
"epoch": 1.6024924795874518,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004927045480340475,
|
|
"loss": 6.3822,
|
|
"mean_token_accuracy": 0.1305439554154873,
|
|
"num_tokens": 3394009.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.446542024612427,
|
|
"epoch": 1.6067898581865063,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004926203315668363,
|
|
"loss": 6.3698,
|
|
"mean_token_accuracy": 0.14797376990318298,
|
|
"num_tokens": 3402550.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.390069961547852,
|
|
"epoch": 1.6110872367855609,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004925356399038032,
|
|
"loss": 6.3408,
|
|
"mean_token_accuracy": 0.1447392597794533,
|
|
"num_tokens": 3413230.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.27685923576355,
|
|
"epoch": 1.6153846153846154,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004924504732298808,
|
|
"loss": 6.3117,
|
|
"mean_token_accuracy": 0.14603087455034255,
|
|
"num_tokens": 3422325.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.474649381637573,
|
|
"epoch": 1.61968199398367,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004923648317310391,
|
|
"loss": 6.392,
|
|
"mean_token_accuracy": 0.13447986543178558,
|
|
"num_tokens": 3431758.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.331472253799438,
|
|
"epoch": 1.6239793725827245,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004922787155942849,
|
|
"loss": 6.2769,
|
|
"mean_token_accuracy": 0.14332059174776077,
|
|
"num_tokens": 3441281.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.405183362960815,
|
|
"epoch": 1.628276751181779,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004921921250076611,
|
|
"loss": 6.32,
|
|
"mean_token_accuracy": 0.1474598653614521,
|
|
"num_tokens": 3449822.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.439214515686035,
|
|
"epoch": 1.6325741297808336,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004921050601602475,
|
|
"loss": 6.3252,
|
|
"mean_token_accuracy": 0.14000287130475045,
|
|
"num_tokens": 3459437.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.364074611663819,
|
|
"epoch": 1.6368715083798882,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004920175212421587,
|
|
"loss": 6.3453,
|
|
"mean_token_accuracy": 0.14289369359612464,
|
|
"num_tokens": 3469546.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.427534103393555,
|
|
"epoch": 1.6411688869789427,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004919295084445445,
|
|
"loss": 6.316,
|
|
"mean_token_accuracy": 0.13860969692468644,
|
|
"num_tokens": 3478521.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.320402002334594,
|
|
"epoch": 1.6454662655779975,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004918410219595899,
|
|
"loss": 6.2135,
|
|
"mean_token_accuracy": 0.1451512724161148,
|
|
"num_tokens": 3487958.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.288850164413452,
|
|
"epoch": 1.649763644177052,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000491752061980514,
|
|
"loss": 6.3142,
|
|
"mean_token_accuracy": 0.14572881534695625,
|
|
"num_tokens": 3495941.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.415158891677857,
|
|
"epoch": 1.6540610227761066,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004916626287015697,
|
|
"loss": 6.3295,
|
|
"mean_token_accuracy": 0.1428305320441723,
|
|
"num_tokens": 3505172.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.391680335998535,
|
|
"epoch": 1.658358401375161,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004915727223180436,
|
|
"loss": 6.3138,
|
|
"mean_token_accuracy": 0.14527018666267394,
|
|
"num_tokens": 3514084.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.442883253097534,
|
|
"epoch": 1.6626557799742159,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004914823430262554,
|
|
"loss": 6.4421,
|
|
"mean_token_accuracy": 0.13630780205130577,
|
|
"num_tokens": 3523318.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.3719299793243405,
|
|
"epoch": 1.6669531585732704,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004913914910235573,
|
|
"loss": 6.2712,
|
|
"mean_token_accuracy": 0.1465964734554291,
|
|
"num_tokens": 3531681.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.30449800491333,
|
|
"epoch": 1.671250537172325,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004913001665083337,
|
|
"loss": 6.3462,
|
|
"mean_token_accuracy": 0.14266296550631524,
|
|
"num_tokens": 3540326.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.393455266952515,
|
|
"epoch": 1.6755479157713795,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004912083696800008,
|
|
"loss": 6.2845,
|
|
"mean_token_accuracy": 0.13957952111959457,
|
|
"num_tokens": 3548319.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.2641520500183105,
|
|
"epoch": 1.679845294370434,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004911161007390063,
|
|
"loss": 6.1394,
|
|
"mean_token_accuracy": 0.15182972550392151,
|
|
"num_tokens": 3557743.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.277702379226684,
|
|
"epoch": 1.6841426729694886,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004910233598868287,
|
|
"loss": 6.2682,
|
|
"mean_token_accuracy": 0.14529897049069404,
|
|
"num_tokens": 3566810.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.337201499938965,
|
|
"epoch": 1.6884400515685432,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004909301473259769,
|
|
"loss": 6.3446,
|
|
"mean_token_accuracy": 0.1441364496946335,
|
|
"num_tokens": 3576273.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.315463352203369,
|
|
"epoch": 1.6927374301675977,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004908364632599899,
|
|
"loss": 6.3265,
|
|
"mean_token_accuracy": 0.1417617380619049,
|
|
"num_tokens": 3585362.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.334482049942016,
|
|
"epoch": 1.6970348087666522,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004907423078934362,
|
|
"loss": 6.3112,
|
|
"mean_token_accuracy": 0.1375837117433548,
|
|
"num_tokens": 3594445.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.350462293624878,
|
|
"epoch": 1.7013321873657068,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004906476814319134,
|
|
"loss": 6.3339,
|
|
"mean_token_accuracy": 0.14064698591828345,
|
|
"num_tokens": 3604046.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.426722812652588,
|
|
"epoch": 1.7056295659647613,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004905525840820481,
|
|
"loss": 6.3385,
|
|
"mean_token_accuracy": 0.14573438987135887,
|
|
"num_tokens": 3613264.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.352614307403565,
|
|
"epoch": 1.709926944563816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004904570160514948,
|
|
"loss": 6.2274,
|
|
"mean_token_accuracy": 0.1472587391734123,
|
|
"num_tokens": 3620938.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.332766199111939,
|
|
"epoch": 1.7142243231628707,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004903609775489358,
|
|
"loss": 6.3316,
|
|
"mean_token_accuracy": 0.141658017039299,
|
|
"num_tokens": 3629782.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.470893049240113,
|
|
"epoch": 1.7185217017619252,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004902644687840809,
|
|
"loss": 6.3851,
|
|
"mean_token_accuracy": 0.13716112747788428,
|
|
"num_tokens": 3638389.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7185217017619252,
|
|
"eval_entropy": 6.099528554323557,
|
|
"eval_loss": 6.394359588623047,
|
|
"eval_mean_token_accuracy": 0.1416532844808456,
|
|
"eval_num_tokens": 3638389.0,
|
|
"eval_runtime": 2.6474,
|
|
"eval_samples_per_second": 1340.554,
|
|
"eval_steps_per_second": 167.711,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.261544561386108,
|
|
"epoch": 1.7228190803609797,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004901674899676667,
|
|
"loss": 6.2108,
|
|
"mean_token_accuracy": 0.15050435364246367,
|
|
"num_tokens": 3647196.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.274995756149292,
|
|
"epoch": 1.7271164589600345,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004900700413114561,
|
|
"loss": 6.1529,
|
|
"mean_token_accuracy": 0.14831772446632385,
|
|
"num_tokens": 3655931.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.3229820728302,
|
|
"epoch": 1.731413837559089,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000489972123028238,
|
|
"loss": 6.321,
|
|
"mean_token_accuracy": 0.14567042738199235,
|
|
"num_tokens": 3665340.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.3659852027893065,
|
|
"epoch": 1.7357112161581436,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004898737353318268,
|
|
"loss": 6.2768,
|
|
"mean_token_accuracy": 0.1477080449461937,
|
|
"num_tokens": 3674674.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.348592519760132,
|
|
"epoch": 1.7400085947571982,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489774878437062,
|
|
"loss": 6.3145,
|
|
"mean_token_accuracy": 0.14947012513875962,
|
|
"num_tokens": 3683910.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.259040975570679,
|
|
"epoch": 1.7443059733562527,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004896755525598074,
|
|
"loss": 6.2112,
|
|
"mean_token_accuracy": 0.14689078107476233,
|
|
"num_tokens": 3692772.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.342519044876099,
|
|
"epoch": 1.7486033519553073,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004895757579169511,
|
|
"loss": 6.2427,
|
|
"mean_token_accuracy": 0.15017979145050048,
|
|
"num_tokens": 3701875.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.292976188659668,
|
|
"epoch": 1.7529007305543618,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004894754947264047,
|
|
"loss": 6.2553,
|
|
"mean_token_accuracy": 0.13835104927420616,
|
|
"num_tokens": 3711059.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.299735260009766,
|
|
"epoch": 1.7571981091534163,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489374763207103,
|
|
"loss": 6.2281,
|
|
"mean_token_accuracy": 0.15146950185298919,
|
|
"num_tokens": 3719667.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.301810169219971,
|
|
"epoch": 1.761495487752471,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004892735635790033,
|
|
"loss": 6.2123,
|
|
"mean_token_accuracy": 0.1495981313288212,
|
|
"num_tokens": 3728905.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.30145845413208,
|
|
"epoch": 1.7657928663515254,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000489171896063085,
|
|
"loss": 6.2537,
|
|
"mean_token_accuracy": 0.1459835670888424,
|
|
"num_tokens": 3737973.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.2883388042449955,
|
|
"epoch": 1.77009024495058,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004890697608813495,
|
|
"loss": 6.2163,
|
|
"mean_token_accuracy": 0.14961224421858788,
|
|
"num_tokens": 3747160.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.263640260696411,
|
|
"epoch": 1.7743876235496348,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004889671582568193,
|
|
"loss": 6.1784,
|
|
"mean_token_accuracy": 0.15031145811080932,
|
|
"num_tokens": 3754908.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.346582460403442,
|
|
"epoch": 1.7786850021486893,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004888640884135374,
|
|
"loss": 6.2829,
|
|
"mean_token_accuracy": 0.13900179788470268,
|
|
"num_tokens": 3763588.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.257235670089722,
|
|
"epoch": 1.7829823807477438,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004887605515765671,
|
|
"loss": 6.1709,
|
|
"mean_token_accuracy": 0.14929060488939286,
|
|
"num_tokens": 3771932.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.365653419494629,
|
|
"epoch": 1.7872797593467986,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004886565479719914,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.14113251492381096,
|
|
"num_tokens": 3780761.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.262348651885986,
|
|
"epoch": 1.7915771379458532,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004885520778269128,
|
|
"loss": 6.1321,
|
|
"mean_token_accuracy": 0.1545582115650177,
|
|
"num_tokens": 3789511.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.2817658424377445,
|
|
"epoch": 1.7958745165449077,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004884471413694523,
|
|
"loss": 6.314,
|
|
"mean_token_accuracy": 0.13621936738491058,
|
|
"num_tokens": 3799068.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.329490661621094,
|
|
"epoch": 1.8001718951439623,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004883417388287491,
|
|
"loss": 6.203,
|
|
"mean_token_accuracy": 0.14342657700181008,
|
|
"num_tokens": 3807850.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.38734540939331,
|
|
"epoch": 1.8044692737430168,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004882358704349603,
|
|
"loss": 6.2334,
|
|
"mean_token_accuracy": 0.1539076805114746,
|
|
"num_tokens": 3816232.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.277137184143067,
|
|
"epoch": 1.8087666523420713,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004881295364192601,
|
|
"loss": 6.2852,
|
|
"mean_token_accuracy": 0.14138692021369934,
|
|
"num_tokens": 3825334.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.2631055355072025,
|
|
"epoch": 1.813064030941126,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004880227370138394,
|
|
"loss": 6.249,
|
|
"mean_token_accuracy": 0.1496968075633049,
|
|
"num_tokens": 3834756.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.390051364898682,
|
|
"epoch": 1.8173614095401804,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004879154724519057,
|
|
"loss": 6.3121,
|
|
"mean_token_accuracy": 0.14131874218583107,
|
|
"num_tokens": 3843852.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.326163721084595,
|
|
"epoch": 1.821658788139235,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004878077429676816,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.15569255948066713,
|
|
"num_tokens": 3852213.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.302305793762207,
|
|
"epoch": 1.8259561667382895,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004876995487964054,
|
|
"loss": 6.3249,
|
|
"mean_token_accuracy": 0.13407231569290162,
|
|
"num_tokens": 3861522.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.268641996383667,
|
|
"epoch": 1.830253545337344,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048759089017432996,
|
|
"loss": 6.2752,
|
|
"mean_token_accuracy": 0.1495513841509819,
|
|
"num_tokens": 3870328.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.366842555999756,
|
|
"epoch": 1.8345509239363988,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004874817673387222,
|
|
"loss": 6.253,
|
|
"mean_token_accuracy": 0.14391155242919923,
|
|
"num_tokens": 3880399.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.331115055084228,
|
|
"epoch": 1.8388483025354534,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048737218052786275,
|
|
"loss": 6.202,
|
|
"mean_token_accuracy": 0.146192529797554,
|
|
"num_tokens": 3889859.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.224690389633179,
|
|
"epoch": 1.843145681134508,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048726212998104554,
|
|
"loss": 6.1712,
|
|
"mean_token_accuracy": 0.1507162019610405,
|
|
"num_tokens": 3899609.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.3443187236785885,
|
|
"epoch": 1.8474430597335625,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004871516159385768,
|
|
"loss": 6.2796,
|
|
"mean_token_accuracy": 0.13646257296204567,
|
|
"num_tokens": 3909452.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.383110523223877,
|
|
"epoch": 1.8517404383326173,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004870406386417752,
|
|
"loss": 6.3188,
|
|
"mean_token_accuracy": 0.14102417081594468,
|
|
"num_tokens": 3918579.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.2510617733001705,
|
|
"epoch": 1.8560378169316718,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004869291983329707,
|
|
"loss": 6.2026,
|
|
"mean_token_accuracy": 0.15068159848451615,
|
|
"num_tokens": 3926951.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.326785755157471,
|
|
"epoch": 1.8603351955307263,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004868172952555044,
|
|
"loss": 6.2538,
|
|
"mean_token_accuracy": 0.14530059248208999,
|
|
"num_tokens": 3935771.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.247272396087647,
|
|
"epoch": 1.864632574129781,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004867049296537278,
|
|
"loss": 6.1507,
|
|
"mean_token_accuracy": 0.15399375408887864,
|
|
"num_tokens": 3944577.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.3375420570373535,
|
|
"epoch": 1.8689299527288354,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004865921017730027,
|
|
"loss": 6.3118,
|
|
"mean_token_accuracy": 0.14250392094254494,
|
|
"num_tokens": 3954499.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.286665630340576,
|
|
"epoch": 1.87322733132789,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048647881185969995,
|
|
"loss": 6.2527,
|
|
"mean_token_accuracy": 0.14288917630910875,
|
|
"num_tokens": 3963953.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.279871034622192,
|
|
"epoch": 1.8775247099269445,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004863650601611994,
|
|
"loss": 6.2466,
|
|
"mean_token_accuracy": 0.14417033568024634,
|
|
"num_tokens": 3973925.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.310920095443725,
|
|
"epoch": 1.881822088525999,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048625084692588937,
|
|
"loss": 6.2296,
|
|
"mean_token_accuracy": 0.1480042040348053,
|
|
"num_tokens": 3983280.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.281775856018067,
|
|
"epoch": 1.8861194671250536,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048613617240316593,
|
|
"loss": 6.2275,
|
|
"mean_token_accuracy": 0.14622583016753196,
|
|
"num_tokens": 3992767.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.342629909515381,
|
|
"epoch": 1.8904168457241082,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004860210368434323,
|
|
"loss": 6.3037,
|
|
"mean_token_accuracy": 0.1325987122952938,
|
|
"num_tokens": 4001600.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.265765047073364,
|
|
"epoch": 1.8947142243231627,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048590544049809857,
|
|
"loss": 6.2767,
|
|
"mean_token_accuracy": 0.1438748598098755,
|
|
"num_tokens": 4010852.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.340087509155273,
|
|
"epoch": 1.8990116029222175,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000485789383619581,
|
|
"loss": 6.1808,
|
|
"mean_token_accuracy": 0.1515198156237602,
|
|
"num_tokens": 4020280.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.251649570465088,
|
|
"epoch": 1.903308981521272,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004856728664613015,
|
|
"loss": 6.1924,
|
|
"mean_token_accuracy": 0.14853738620877266,
|
|
"num_tokens": 4029283.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.3928605079650875,
|
|
"epoch": 1.9076063601203266,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048555588927768674,
|
|
"loss": 6.2813,
|
|
"mean_token_accuracy": 0.15265752896666526,
|
|
"num_tokens": 4038704.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.133363246917725,
|
|
"epoch": 1.9119037387193811,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004854384523241683,
|
|
"loss": 6.149,
|
|
"mean_token_accuracy": 0.14911998957395553,
|
|
"num_tokens": 4047492.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.350856971740723,
|
|
"epoch": 1.916201117318436,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048532055585718143,
|
|
"loss": 6.3258,
|
|
"mean_token_accuracy": 0.14355307817459106,
|
|
"num_tokens": 4056792.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.365505075454712,
|
|
"epoch": 1.9204984959174904,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048520220013416505,
|
|
"loss": 6.2246,
|
|
"mean_token_accuracy": 0.1529555104672909,
|
|
"num_tokens": 4065284.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.168982744216919,
|
|
"epoch": 1.924795874516545,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004850833854135607,
|
|
"loss": 6.2088,
|
|
"mean_token_accuracy": 0.1461514100432396,
|
|
"num_tokens": 4074727.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.41944990158081,
|
|
"epoch": 1.9290932531155995,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004849641119548122,
|
|
"loss": 6.2592,
|
|
"mean_token_accuracy": 0.15032583549618722,
|
|
"num_tokens": 4083360.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.2758283615112305,
|
|
"epoch": 1.933390631714654,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000484844380018365,
|
|
"loss": 6.3513,
|
|
"mean_token_accuracy": 0.13633401170372964,
|
|
"num_tokens": 4092431.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.3127405643463135,
|
|
"epoch": 1.9376880103137086,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000484724189865666,
|
|
"loss": 6.2257,
|
|
"mean_token_accuracy": 0.1493070751428604,
|
|
"num_tokens": 4102198.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.309014892578125,
|
|
"epoch": 1.9419853889127632,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004846035417591624,
|
|
"loss": 6.2159,
|
|
"mean_token_accuracy": 0.14939168095588684,
|
|
"num_tokens": 4111568.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.260059499740601,
|
|
"epoch": 1.9462827675118177,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004844824359623014,
|
|
"loss": 6.2428,
|
|
"mean_token_accuracy": 0.1440415069460869,
|
|
"num_tokens": 4121151.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.318060827255249,
|
|
"epoch": 1.9505801461108723,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048436087273952966,
|
|
"loss": 6.1757,
|
|
"mean_token_accuracy": 0.14666527807712554,
|
|
"num_tokens": 4130197.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.161466360092163,
|
|
"epoch": 1.9548775247099268,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048423885235629265,
|
|
"loss": 6.1784,
|
|
"mean_token_accuracy": 0.1395201563835144,
|
|
"num_tokens": 4138310.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.2752457618713375,
|
|
"epoch": 1.9591749033089814,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004841163750790342,
|
|
"loss": 6.1845,
|
|
"mean_token_accuracy": 0.15314925760030745,
|
|
"num_tokens": 4148270.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.315674591064453,
|
|
"epoch": 1.9634722819080361,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048399344117519555,
|
|
"loss": 6.2468,
|
|
"mean_token_accuracy": 0.14707700312137603,
|
|
"num_tokens": 4157276.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.204446601867676,
|
|
"epoch": 1.9677696605070907,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048387005091321544,
|
|
"loss": 6.2742,
|
|
"mean_token_accuracy": 0.15154237672686577,
|
|
"num_tokens": 4166144.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.408171319961548,
|
|
"epoch": 1.9720670391061452,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048374620456252877,
|
|
"loss": 6.2501,
|
|
"mean_token_accuracy": 0.14817840680480004,
|
|
"num_tokens": 4175146.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.269484567642212,
|
|
"epoch": 1.9763644177052,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048362190239356644,
|
|
"loss": 6.2973,
|
|
"mean_token_accuracy": 0.14276304692029954,
|
|
"num_tokens": 4184089.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.2955399513244625,
|
|
"epoch": 1.9806617963042545,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048349714467775474,
|
|
"loss": 6.2199,
|
|
"mean_token_accuracy": 0.14973291605710984,
|
|
"num_tokens": 4192453.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.279147005081176,
|
|
"epoch": 1.984959174903309,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048337193168751464,
|
|
"loss": 6.1458,
|
|
"mean_token_accuracy": 0.15528891682624818,
|
|
"num_tokens": 4201323.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.252904844284058,
|
|
"epoch": 1.9892565535023636,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004832462636962613,
|
|
"loss": 6.1965,
|
|
"mean_token_accuracy": 0.14820378869771958,
|
|
"num_tokens": 4209805.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.274013090133667,
|
|
"epoch": 1.9935539321014182,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004831201409784034,
|
|
"loss": 6.1716,
|
|
"mean_token_accuracy": 0.1535733938217163,
|
|
"num_tokens": 4218562.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.276880836486816,
|
|
"epoch": 1.9978513107004727,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004829935638093424,
|
|
"loss": 6.2489,
|
|
"mean_token_accuracy": 0.14822545498609543,
|
|
"num_tokens": 4227605.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.305307547251384,
|
|
"epoch": 2.0017189514396216,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004828665324654724,
|
|
"loss": 6.0808,
|
|
"mean_token_accuracy": 0.15109746985965306,
|
|
"num_tokens": 4235441.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.1959953784942625,
|
|
"epoch": 2.006016330038676,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004827390472241791,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.15755202025175094,
|
|
"num_tokens": 4243982.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.151334667205811,
|
|
"epoch": 2.010313708637731,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004826111083638392,
|
|
"loss": 5.9654,
|
|
"mean_token_accuracy": 0.15486665964126586,
|
|
"num_tokens": 4253032.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.237190914154053,
|
|
"epoch": 2.0146110872367857,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048248271616382,
|
|
"loss": 5.8539,
|
|
"mean_token_accuracy": 0.1612431138753891,
|
|
"num_tokens": 4262000.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.14679217338562,
|
|
"epoch": 2.0189084658358403,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048235387090447894,
|
|
"loss": 5.8772,
|
|
"mean_token_accuracy": 0.16036980897188186,
|
|
"num_tokens": 4271089.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.216695976257324,
|
|
"epoch": 2.023205844434895,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048222457286716235,
|
|
"loss": 5.8748,
|
|
"mean_token_accuracy": 0.1604112282395363,
|
|
"num_tokens": 4280064.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.1386841297149655,
|
|
"epoch": 2.0275032230339494,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048209482233420564,
|
|
"loss": 5.8701,
|
|
"mean_token_accuracy": 0.16113310903310776,
|
|
"num_tokens": 4289096.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.134747457504273,
|
|
"epoch": 2.031800601633004,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000481964619588932,
|
|
"loss": 5.878,
|
|
"mean_token_accuracy": 0.1504762887954712,
|
|
"num_tokens": 4298341.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.1863165378570555,
|
|
"epoch": 2.0360979802320585,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004818339649156523,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.1504568099975586,
|
|
"num_tokens": 4307466.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.122588872909546,
|
|
"epoch": 2.040395358831113,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048170285859966395,
|
|
"loss": 5.8523,
|
|
"mean_token_accuracy": 0.15609132051467894,
|
|
"num_tokens": 4316559.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.180559778213501,
|
|
"epoch": 2.0446927374301676,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048157130092725087,
|
|
"loss": 5.8607,
|
|
"mean_token_accuracy": 0.1605754405260086,
|
|
"num_tokens": 4326002.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.1395592212677,
|
|
"epoch": 2.048990116029222,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004814392921856824,
|
|
"loss": 5.8467,
|
|
"mean_token_accuracy": 0.16454231590032578,
|
|
"num_tokens": 4333767.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.13576455116272,
|
|
"epoch": 2.0532874946282766,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004813068326632128,
|
|
"loss": 5.8941,
|
|
"mean_token_accuracy": 0.16491636782884597,
|
|
"num_tokens": 4342726.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.0936521053314205,
|
|
"epoch": 2.057584873227331,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004811739226490809,
|
|
"loss": 5.7842,
|
|
"mean_token_accuracy": 0.16501135975122452,
|
|
"num_tokens": 4351278.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.10044412612915,
|
|
"epoch": 2.0618822518263857,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048104056243350896,
|
|
"loss": 5.8079,
|
|
"mean_token_accuracy": 0.16724471151828765,
|
|
"num_tokens": 4360146.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.106970643997192,
|
|
"epoch": 2.0661796304254403,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004809067523077023,
|
|
"loss": 5.8729,
|
|
"mean_token_accuracy": 0.157526695728302,
|
|
"num_tokens": 4368474.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.171187162399292,
|
|
"epoch": 2.0704770090244953,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00048077249256384884,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.1573039174079895,
|
|
"num_tokens": 4378430.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.091280841827393,
|
|
"epoch": 2.07477438762355,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004806377834951182,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.16114620715379716,
|
|
"num_tokens": 4387661.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.2243939399719235,
|
|
"epoch": 2.0790717662226044,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048050262539566104,
|
|
"loss": 5.878,
|
|
"mean_token_accuracy": 0.16368078142404557,
|
|
"num_tokens": 4396568.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.03709397315979,
|
|
"epoch": 2.083369144821659,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004803670185606087,
|
|
"loss": 5.8322,
|
|
"mean_token_accuracy": 0.16086018234491348,
|
|
"num_tokens": 4405113.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.130343770980835,
|
|
"epoch": 2.0876665234207135,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004802309632860724,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.15909243375062943,
|
|
"num_tokens": 4414930.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.135715103149414,
|
|
"epoch": 2.091963902019768,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048009445986914236,
|
|
"loss": 5.9472,
|
|
"mean_token_accuracy": 0.15476218312978746,
|
|
"num_tokens": 4424200.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.087995195388794,
|
|
"epoch": 2.0962612806188226,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047995750860788756,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.16245823204517365,
|
|
"num_tokens": 4433533.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.122352218627929,
|
|
"epoch": 2.100558659217877,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004798201098013547,
|
|
"loss": 5.9404,
|
|
"mean_token_accuracy": 0.1550127923488617,
|
|
"num_tokens": 4442752.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.078053426742554,
|
|
"epoch": 2.1048560378169316,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047968226374956797,
|
|
"loss": 5.971,
|
|
"mean_token_accuracy": 0.15189971923828124,
|
|
"num_tokens": 4453624.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.135152006149292,
|
|
"epoch": 2.109153416415986,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047954397075352794,
|
|
"loss": 5.9528,
|
|
"mean_token_accuracy": 0.15271927714347838,
|
|
"num_tokens": 4462101.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.186189222335815,
|
|
"epoch": 2.1134507950150407,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047940523111521136,
|
|
"loss": 5.9668,
|
|
"mean_token_accuracy": 0.15061386451125144,
|
|
"num_tokens": 4471274.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.176623678207397,
|
|
"epoch": 2.1177481736140953,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004792660451375701,
|
|
"loss": 5.9378,
|
|
"mean_token_accuracy": 0.15836917012929916,
|
|
"num_tokens": 4480281.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.131458759307861,
|
|
"epoch": 2.12204555221315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047912641312453064,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.15682227164506912,
|
|
"num_tokens": 4489325.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.136132478713989,
|
|
"epoch": 2.1263429308122044,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047898633538099363,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.16200080662965774,
|
|
"num_tokens": 4497977.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.082278108596801,
|
|
"epoch": 2.130640309411259,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004788458122128327,
|
|
"loss": 5.9842,
|
|
"mean_token_accuracy": 0.15614993274211883,
|
|
"num_tokens": 4509133.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.15927677154541,
|
|
"epoch": 2.134937688010314,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047870484392689434,
|
|
"loss": 5.9287,
|
|
"mean_token_accuracy": 0.16319320052862168,
|
|
"num_tokens": 4518431.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.090216445922851,
|
|
"epoch": 2.1392350666093685,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000478563430830997,
|
|
"loss": 5.9317,
|
|
"mean_token_accuracy": 0.15791197270154952,
|
|
"num_tokens": 4528489.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.161773586273194,
|
|
"epoch": 2.143532445208423,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047842157323393035,
|
|
"loss": 5.821,
|
|
"mean_token_accuracy": 0.1565369412302971,
|
|
"num_tokens": 4537068.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.096503257751465,
|
|
"epoch": 2.1478298238074776,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004782792714454547,
|
|
"loss": 5.9322,
|
|
"mean_token_accuracy": 0.16345749497413636,
|
|
"num_tokens": 4545773.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.1478298238074776,
|
|
"eval_entropy": 5.961080405089232,
|
|
"eval_loss": 6.284891128540039,
|
|
"eval_mean_token_accuracy": 0.152339708798372,
|
|
"eval_num_tokens": 4545773.0,
|
|
"eval_runtime": 2.6434,
|
|
"eval_samples_per_second": 1342.578,
|
|
"eval_steps_per_second": 167.964,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.136997556686401,
|
|
"epoch": 2.152127202406532,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004781365257763002,
|
|
"loss": 5.9448,
|
|
"mean_token_accuracy": 0.15878659188747407,
|
|
"num_tokens": 4555358.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.074670028686524,
|
|
"epoch": 2.1564245810055866,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047799333653816633,
|
|
"loss": 5.8014,
|
|
"mean_token_accuracy": 0.16225225627422332,
|
|
"num_tokens": 4563550.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.139217233657837,
|
|
"epoch": 2.160721959604641,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047784970404372124,
|
|
"loss": 5.9112,
|
|
"mean_token_accuracy": 0.15919663310050963,
|
|
"num_tokens": 4572517.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.079522275924683,
|
|
"epoch": 2.1650193382036957,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047770562860660083,
|
|
"loss": 5.8478,
|
|
"mean_token_accuracy": 0.16091176122426987,
|
|
"num_tokens": 4581611.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.072164392471313,
|
|
"epoch": 2.1693167168027503,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004775611105414083,
|
|
"loss": 5.8643,
|
|
"mean_token_accuracy": 0.16704804301261902,
|
|
"num_tokens": 4590433.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.130130338668823,
|
|
"epoch": 2.173614095401805,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004774161501637133,
|
|
"loss": 5.9815,
|
|
"mean_token_accuracy": 0.1534938856959343,
|
|
"num_tokens": 4599363.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.083512544631958,
|
|
"epoch": 2.1779114740008594,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004772707477900514,
|
|
"loss": 5.9121,
|
|
"mean_token_accuracy": 0.16265445053577424,
|
|
"num_tokens": 4608647.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.177105474472046,
|
|
"epoch": 2.182208852599914,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004771249037379232,
|
|
"loss": 5.8736,
|
|
"mean_token_accuracy": 0.15716385543346406,
|
|
"num_tokens": 4617395.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.112827777862549,
|
|
"epoch": 2.1865062311989685,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004769786183257939,
|
|
"loss": 5.9587,
|
|
"mean_token_accuracy": 0.1538214087486267,
|
|
"num_tokens": 4627905.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.12396821975708,
|
|
"epoch": 2.190803609798023,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004768318918730924,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.15287947952747344,
|
|
"num_tokens": 4636679.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.11383900642395,
|
|
"epoch": 2.195100988397078,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047668472470021044,
|
|
"loss": 5.9125,
|
|
"mean_token_accuracy": 0.1615729160606861,
|
|
"num_tokens": 4645963.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.046568727493286,
|
|
"epoch": 2.1993983669961326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004765371171285025,
|
|
"loss": 5.8857,
|
|
"mean_token_accuracy": 0.1614224225282669,
|
|
"num_tokens": 4656335.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.01671781539917,
|
|
"epoch": 2.203695745595187,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047638906948028445,
|
|
"loss": 5.8409,
|
|
"mean_token_accuracy": 0.16636864989995956,
|
|
"num_tokens": 4664924.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.137506341934204,
|
|
"epoch": 2.2079931241942417,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047624058207883317,
|
|
"loss": 5.904,
|
|
"mean_token_accuracy": 0.1609740674495697,
|
|
"num_tokens": 4674411.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.012920904159546,
|
|
"epoch": 2.212290502793296,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047609165524838576,
|
|
"loss": 5.8554,
|
|
"mean_token_accuracy": 0.16097129881381989,
|
|
"num_tokens": 4682856.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.120091915130615,
|
|
"epoch": 2.2165878813923507,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004759422893141389,
|
|
"loss": 5.9156,
|
|
"mean_token_accuracy": 0.15652135163545608,
|
|
"num_tokens": 4691652.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.083839035034179,
|
|
"epoch": 2.2208852599914053,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004757924846022482,
|
|
"loss": 5.8458,
|
|
"mean_token_accuracy": 0.1679081380367279,
|
|
"num_tokens": 4700934.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.091445064544677,
|
|
"epoch": 2.22518263859046,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047564224143982714,
|
|
"loss": 5.919,
|
|
"mean_token_accuracy": 0.16310430467128753,
|
|
"num_tokens": 4709195.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.07176628112793,
|
|
"epoch": 2.2294800171895144,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047549156015494676,
|
|
"loss": 6.0021,
|
|
"mean_token_accuracy": 0.1592251941561699,
|
|
"num_tokens": 4719153.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.2358345031738285,
|
|
"epoch": 2.233777395788569,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047534044107663484,
|
|
"loss": 5.9354,
|
|
"mean_token_accuracy": 0.15806687176227568,
|
|
"num_tokens": 4727672.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.045652770996094,
|
|
"epoch": 2.2380747743876235,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047518888453487496,
|
|
"loss": 5.8695,
|
|
"mean_token_accuracy": 0.1585845544934273,
|
|
"num_tokens": 4735893.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.075984144210816,
|
|
"epoch": 2.242372152986678,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004750368908606061,
|
|
"loss": 5.8889,
|
|
"mean_token_accuracy": 0.15606335699558258,
|
|
"num_tokens": 4744022.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.096569538116455,
|
|
"epoch": 2.2466695315857326,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047488446038572164,
|
|
"loss": 5.9698,
|
|
"mean_token_accuracy": 0.1579223982989788,
|
|
"num_tokens": 4752996.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.137444591522216,
|
|
"epoch": 2.250966910184787,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004747315934430688,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.16073836088180543,
|
|
"num_tokens": 4762116.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.994189739227295,
|
|
"epoch": 2.2552642887838417,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000474578290366448,
|
|
"loss": 5.8926,
|
|
"mean_token_accuracy": 0.1578589841723442,
|
|
"num_tokens": 4771154.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.078905153274536,
|
|
"epoch": 2.259561667382896,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004744245514906117,
|
|
"loss": 5.9424,
|
|
"mean_token_accuracy": 0.15930399894714356,
|
|
"num_tokens": 4779691.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.151353120803833,
|
|
"epoch": 2.263859045981951,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047427037715126426,
|
|
"loss": 5.8905,
|
|
"mean_token_accuracy": 0.16179440766572953,
|
|
"num_tokens": 4788033.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.0782910823822025,
|
|
"epoch": 2.2681564245810057,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004741157676850608,
|
|
"loss": 5.8832,
|
|
"mean_token_accuracy": 0.15863569974899291,
|
|
"num_tokens": 4797495.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.083225727081299,
|
|
"epoch": 2.2724538031800603,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047396072342960663,
|
|
"loss": 5.9558,
|
|
"mean_token_accuracy": 0.14972878918051719,
|
|
"num_tokens": 4806598.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.104167509078979,
|
|
"epoch": 2.276751181779115,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047380524472345645,
|
|
"loss": 5.9232,
|
|
"mean_token_accuracy": 0.15156156048178673,
|
|
"num_tokens": 4816135.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.095362854003906,
|
|
"epoch": 2.2810485603781694,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004736493319061134,
|
|
"loss": 5.9192,
|
|
"mean_token_accuracy": 0.15769523978233338,
|
|
"num_tokens": 4825927.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.085659694671631,
|
|
"epoch": 2.285345938977224,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004734929853180291,
|
|
"loss": 5.8491,
|
|
"mean_token_accuracy": 0.15650810301303864,
|
|
"num_tokens": 4834068.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.085105562210083,
|
|
"epoch": 2.2896433175762785,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047333620530060175,
|
|
"loss": 5.844,
|
|
"mean_token_accuracy": 0.16463817805051803,
|
|
"num_tokens": 4843096.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.99607744216919,
|
|
"epoch": 2.293940696175333,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004731789921961764,
|
|
"loss": 5.8948,
|
|
"mean_token_accuracy": 0.16113691926002502,
|
|
"num_tokens": 4850948.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.157177257537842,
|
|
"epoch": 2.2982380747743876,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004730213463480434,
|
|
"loss": 6.0074,
|
|
"mean_token_accuracy": 0.15841376781463623,
|
|
"num_tokens": 4860416.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.064683675765991,
|
|
"epoch": 2.302535453373442,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047286326810043857,
|
|
"loss": 5.8577,
|
|
"mean_token_accuracy": 0.16336074024438857,
|
|
"num_tokens": 4868815.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.055120992660522,
|
|
"epoch": 2.3068328319724967,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047270475779854137,
|
|
"loss": 5.835,
|
|
"mean_token_accuracy": 0.15979370176792146,
|
|
"num_tokens": 4877973.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.094270849227906,
|
|
"epoch": 2.311130210571551,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047254581578847507,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.168157596886158,
|
|
"num_tokens": 4887294.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.958069801330566,
|
|
"epoch": 2.3154275891706058,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004723864424173055,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.16242188811302186,
|
|
"num_tokens": 4895618.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.042819690704346,
|
|
"epoch": 2.3197249677696608,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004722266380330403,
|
|
"loss": 5.8829,
|
|
"mean_token_accuracy": 0.15943997576832772,
|
|
"num_tokens": 4904702.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.096137475967407,
|
|
"epoch": 2.3240223463687153,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047206640298462857,
|
|
"loss": 5.9278,
|
|
"mean_token_accuracy": 0.1653011530637741,
|
|
"num_tokens": 4913677.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.084006643295288,
|
|
"epoch": 2.32831972496777,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047190573762195945,
|
|
"loss": 6.0771,
|
|
"mean_token_accuracy": 0.14496880918741226,
|
|
"num_tokens": 4924192.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.091192293167114,
|
|
"epoch": 2.3326171035668244,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047174464229586186,
|
|
"loss": 5.788,
|
|
"mean_token_accuracy": 0.16894155740737915,
|
|
"num_tokens": 4933424.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.024014091491699,
|
|
"epoch": 2.336914482165879,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004715831173581036,
|
|
"loss": 5.9315,
|
|
"mean_token_accuracy": 0.16265107691287994,
|
|
"num_tokens": 4942695.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.095777130126953,
|
|
"epoch": 2.3412118607649335,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047142116316139073,
|
|
"loss": 6.0217,
|
|
"mean_token_accuracy": 0.1556411489844322,
|
|
"num_tokens": 4952951.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.034127712249756,
|
|
"epoch": 2.345509239363988,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004712587800593663,
|
|
"loss": 5.8689,
|
|
"mean_token_accuracy": 0.16399069130420685,
|
|
"num_tokens": 4961750.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.052180910110474,
|
|
"epoch": 2.3498066179630426,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004710959684066102,
|
|
"loss": 5.975,
|
|
"mean_token_accuracy": 0.1578183650970459,
|
|
"num_tokens": 4970999.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.984068822860718,
|
|
"epoch": 2.354103996562097,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047093272855863803,
|
|
"loss": 5.7811,
|
|
"mean_token_accuracy": 0.1679188758134842,
|
|
"num_tokens": 4979703.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.053386402130127,
|
|
"epoch": 2.3584013751611517,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004707690608719003,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.16708777993917465,
|
|
"num_tokens": 4988666.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.0471906661987305,
|
|
"epoch": 2.362698753760206,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004706049657037818,
|
|
"loss": 5.8996,
|
|
"mean_token_accuracy": 0.15584399104118346,
|
|
"num_tokens": 4997741.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.02504529953003,
|
|
"epoch": 2.3669961323592608,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0004704404434126009,
|
|
"loss": 5.8461,
|
|
"mean_token_accuracy": 0.15902445018291472,
|
|
"num_tokens": 5006067.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.009567022323608,
|
|
"epoch": 2.3712935109583153,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047027549435760843,
|
|
"loss": 5.8469,
|
|
"mean_token_accuracy": 0.16029683277010917,
|
|
"num_tokens": 5015212.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.1162244319915775,
|
|
"epoch": 2.37559088955737,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004701101188989872,
|
|
"loss": 5.9369,
|
|
"mean_token_accuracy": 0.15840114057064056,
|
|
"num_tokens": 5025461.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.126107692718506,
|
|
"epoch": 2.3798882681564244,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00046994431739785114,
|
|
"loss": 5.9264,
|
|
"mean_token_accuracy": 0.15688806027173996,
|
|
"num_tokens": 5033479.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.02915506362915,
|
|
"epoch": 2.384185646755479,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046977809021624454,
|
|
"loss": 5.9723,
|
|
"mean_token_accuracy": 0.15953245013952255,
|
|
"num_tokens": 5043098.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.0680704593658445,
|
|
"epoch": 2.3884830253545335,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004696114377171409,
|
|
"loss": 5.9199,
|
|
"mean_token_accuracy": 0.16611011028289796,
|
|
"num_tokens": 5052785.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.041296768188476,
|
|
"epoch": 2.3927804039535885,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004694443602644429,
|
|
"loss": 5.875,
|
|
"mean_token_accuracy": 0.16046594679355622,
|
|
"num_tokens": 5061947.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.086552667617798,
|
|
"epoch": 2.397077782552643,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004692768582229808,
|
|
"loss": 5.8978,
|
|
"mean_token_accuracy": 0.1659877672791481,
|
|
"num_tokens": 5070624.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.069063282012939,
|
|
"epoch": 2.4013751611516976,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00046910893195851213,
|
|
"loss": 5.9253,
|
|
"mean_token_accuracy": 0.1656736984848976,
|
|
"num_tokens": 5079290.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.013190841674804,
|
|
"epoch": 2.405672539750752,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046894058183772074,
|
|
"loss": 5.8435,
|
|
"mean_token_accuracy": 0.15861762911081315,
|
|
"num_tokens": 5088214.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.016868448257446,
|
|
"epoch": 2.4099699183498067,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000468771808228216,
|
|
"loss": 5.8659,
|
|
"mean_token_accuracy": 0.15881743729114534,
|
|
"num_tokens": 5096752.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.110416555404663,
|
|
"epoch": 2.414267296948861,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046860261149853197,
|
|
"loss": 5.9618,
|
|
"mean_token_accuracy": 0.15770885795354844,
|
|
"num_tokens": 5105204.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.0625450134277346,
|
|
"epoch": 2.4185646755479158,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004684329920181268,
|
|
"loss": 5.8465,
|
|
"mean_token_accuracy": 0.16744299978017807,
|
|
"num_tokens": 5114306.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.018719100952149,
|
|
"epoch": 2.4228620541469703,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046826295015738154,
|
|
"loss": 5.8693,
|
|
"mean_token_accuracy": 0.1659521371126175,
|
|
"num_tokens": 5122894.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.021153783798217,
|
|
"epoch": 2.427159432746025,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004680924862875996,
|
|
"loss": 5.9744,
|
|
"mean_token_accuracy": 0.16065433025360107,
|
|
"num_tokens": 5131660.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.128577804565429,
|
|
"epoch": 2.4314568113450794,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046792160078100605,
|
|
"loss": 5.9232,
|
|
"mean_token_accuracy": 0.15939103215932846,
|
|
"num_tokens": 5140165.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.055081415176391,
|
|
"epoch": 2.435754189944134,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046775029401074653,
|
|
"loss": 5.8907,
|
|
"mean_token_accuracy": 0.1634244441986084,
|
|
"num_tokens": 5149998.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.023366022109985,
|
|
"epoch": 2.4400515685431885,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00046757856635088645,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.16400753408670427,
|
|
"num_tokens": 5159182.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.150636291503906,
|
|
"epoch": 2.444348947142243,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004674064181764105,
|
|
"loss": 6.026,
|
|
"mean_token_accuracy": 0.14548912644386292,
|
|
"num_tokens": 5169087.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.079051303863525,
|
|
"epoch": 2.448646325741298,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046723384986322147,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.1683396190404892,
|
|
"num_tokens": 5177670.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.105928993225097,
|
|
"epoch": 2.4529437043403526,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004670608617881395,
|
|
"loss": 6.0178,
|
|
"mean_token_accuracy": 0.15267052501440048,
|
|
"num_tokens": 5187273.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.084513902664185,
|
|
"epoch": 2.457241082939407,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004668874543289014,
|
|
"loss": 5.8832,
|
|
"mean_token_accuracy": 0.1608045980334282,
|
|
"num_tokens": 5196546.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.099105167388916,
|
|
"epoch": 2.4615384615384617,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046671362786415986,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.16351703703403472,
|
|
"num_tokens": 5205966.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.989667177200317,
|
|
"epoch": 2.465835840137516,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00046653938277348237,
|
|
"loss": 5.9724,
|
|
"mean_token_accuracy": 0.16026378571987152,
|
|
"num_tokens": 5216042.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.0985715866088865,
|
|
"epoch": 2.4701332187365708,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004663647194373505,
|
|
"loss": 5.955,
|
|
"mean_token_accuracy": 0.1608792155981064,
|
|
"num_tokens": 5225514.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.063877964019776,
|
|
"epoch": 2.4744305973356253,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046618963823715913,
|
|
"loss": 5.8931,
|
|
"mean_token_accuracy": 0.1574872240424156,
|
|
"num_tokens": 5233705.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.085922336578369,
|
|
"epoch": 2.47872797593468,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046601413955521575,
|
|
"loss": 5.9292,
|
|
"mean_token_accuracy": 0.1562924087047577,
|
|
"num_tokens": 5242820.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.060844755172729,
|
|
"epoch": 2.4830253545337344,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004658382237747393,
|
|
"loss": 5.9735,
|
|
"mean_token_accuracy": 0.16258694231510162,
|
|
"num_tokens": 5252036.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.9974912166595455,
|
|
"epoch": 2.487322733132789,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046566189127985946,
|
|
"loss": 5.8077,
|
|
"mean_token_accuracy": 0.1622447907924652,
|
|
"num_tokens": 5260734.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.083310508728028,
|
|
"epoch": 2.4916201117318435,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000465485142455616,
|
|
"loss": 5.9405,
|
|
"mean_token_accuracy": 0.15838106870651245,
|
|
"num_tokens": 5269764.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.100842475891113,
|
|
"epoch": 2.495917490330898,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046530797768795765,
|
|
"loss": 5.9003,
|
|
"mean_token_accuracy": 0.16507043689489365,
|
|
"num_tokens": 5278585.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.027634620666504,
|
|
"epoch": 2.5002148689299526,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046513039736374153,
|
|
"loss": 5.9417,
|
|
"mean_token_accuracy": 0.15948257148265838,
|
|
"num_tokens": 5287335.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.155354166030884,
|
|
"epoch": 2.504512247529007,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004649524018707319,
|
|
"loss": 6.0748,
|
|
"mean_token_accuracy": 0.14517340958118438,
|
|
"num_tokens": 5297977.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.119415283203125,
|
|
"epoch": 2.5088096261280617,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046477399159759996,
|
|
"loss": 5.9254,
|
|
"mean_token_accuracy": 0.15665082335472108,
|
|
"num_tokens": 5307148.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.057867956161499,
|
|
"epoch": 2.5131070047271162,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046459516693392246,
|
|
"loss": 5.9712,
|
|
"mean_token_accuracy": 0.15419441908597947,
|
|
"num_tokens": 5317046.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.140188217163086,
|
|
"epoch": 2.517404383326171,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004644159282701808,
|
|
"loss": 5.9725,
|
|
"mean_token_accuracy": 0.16102391481399536,
|
|
"num_tokens": 5325978.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.035956478118896,
|
|
"epoch": 2.5217017619252258,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046423627599776076,
|
|
"loss": 5.8578,
|
|
"mean_token_accuracy": 0.16485689729452133,
|
|
"num_tokens": 5334804.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.078884077072144,
|
|
"epoch": 2.5259991405242803,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000464056210508951,
|
|
"loss": 5.8277,
|
|
"mean_token_accuracy": 0.16394184082746505,
|
|
"num_tokens": 5343575.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.003749418258667,
|
|
"epoch": 2.530296519123335,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004638757321969426,
|
|
"loss": 5.8464,
|
|
"mean_token_accuracy": 0.16426144391298295,
|
|
"num_tokens": 5352151.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.079893970489502,
|
|
"epoch": 2.5345938977223894,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046369484145582815,
|
|
"loss": 6.0248,
|
|
"mean_token_accuracy": 0.1487678438425064,
|
|
"num_tokens": 5363048.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.128987884521484,
|
|
"epoch": 2.538891276321444,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046351353868060054,
|
|
"loss": 5.9391,
|
|
"mean_token_accuracy": 0.15860069245100022,
|
|
"num_tokens": 5372365.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.0094362735748295,
|
|
"epoch": 2.5431886549204985,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00046333182426715273,
|
|
"loss": 5.902,
|
|
"mean_token_accuracy": 0.16009646207094191,
|
|
"num_tokens": 5381583.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.9937304019927975,
|
|
"epoch": 2.547486033519553,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046314969861227626,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.15933951884508132,
|
|
"num_tokens": 5390746.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.056115436553955,
|
|
"epoch": 2.5517834121186076,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004629671621136608,
|
|
"loss": 5.9144,
|
|
"mean_token_accuracy": 0.15412782430648803,
|
|
"num_tokens": 5399715.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.087065887451172,
|
|
"epoch": 2.556080790717662,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004627842151698931,
|
|
"loss": 5.9459,
|
|
"mean_token_accuracy": 0.1620703786611557,
|
|
"num_tokens": 5407906.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.994268703460693,
|
|
"epoch": 2.5603781693167167,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00046260085818045625,
|
|
"loss": 5.8688,
|
|
"mean_token_accuracy": 0.16364364624023436,
|
|
"num_tokens": 5416935.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.079984045028686,
|
|
"epoch": 2.5646755479157712,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004624170915457284,
|
|
"loss": 5.9326,
|
|
"mean_token_accuracy": 0.16039495617151261,
|
|
"num_tokens": 5426061.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.036399412155151,
|
|
"epoch": 2.5689729265148262,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046223291566698264,
|
|
"loss": 5.9213,
|
|
"mean_token_accuracy": 0.1574646696448326,
|
|
"num_tokens": 5435546.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.026375961303711,
|
|
"epoch": 2.5732703051138808,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004620483309463855,
|
|
"loss": 5.8993,
|
|
"mean_token_accuracy": 0.1682535871863365,
|
|
"num_tokens": 5445026.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.025810480117798,
|
|
"epoch": 2.5775676837129353,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004618633377869961,
|
|
"loss": 5.8568,
|
|
"mean_token_accuracy": 0.16291579902172088,
|
|
"num_tokens": 5453936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.5775676837129353,
|
|
"eval_entropy": 6.043743693076812,
|
|
"eval_loss": 6.22243070602417,
|
|
"eval_mean_token_accuracy": 0.15526222028173842,
|
|
"eval_num_tokens": 5453936.0,
|
|
"eval_runtime": 2.6441,
|
|
"eval_samples_per_second": 1342.222,
|
|
"eval_steps_per_second": 167.92,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.1455272197723385,
|
|
"epoch": 2.58186506231199,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004616779365927656,
|
|
"loss": 5.9191,
|
|
"mean_token_accuracy": 0.17172190099954604,
|
|
"num_tokens": 5462313.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.955387449264526,
|
|
"epoch": 2.5861624409110444,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004614921277685361,
|
|
"loss": 5.8934,
|
|
"mean_token_accuracy": 0.16470066010951995,
|
|
"num_tokens": 5471500.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.011490392684936,
|
|
"epoch": 2.590459819510099,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00046130591172003976,
|
|
"loss": 5.8353,
|
|
"mean_token_accuracy": 0.16367537677288055,
|
|
"num_tokens": 5480317.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.021433734893799,
|
|
"epoch": 2.5947571981091535,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004611192888538981,
|
|
"loss": 5.8961,
|
|
"mean_token_accuracy": 0.1588372528553009,
|
|
"num_tokens": 5490159.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.054172992706299,
|
|
"epoch": 2.599054576708208,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00046093225957762084,
|
|
"loss": 5.8504,
|
|
"mean_token_accuracy": 0.17010194659233094,
|
|
"num_tokens": 5499055.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.099794387817383,
|
|
"epoch": 2.6033519553072626,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004607448242996051,
|
|
"loss": 5.9703,
|
|
"mean_token_accuracy": 0.1583339735865593,
|
|
"num_tokens": 5508710.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.093514060974121,
|
|
"epoch": 2.607649333906317,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004605569834291347,
|
|
"loss": 5.9042,
|
|
"mean_token_accuracy": 0.1637094348669052,
|
|
"num_tokens": 5517060.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.978723049163818,
|
|
"epoch": 2.6119467125053717,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046036873737637904,
|
|
"loss": 5.8248,
|
|
"mean_token_accuracy": 0.1596076935529709,
|
|
"num_tokens": 5525652.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.008916664123535,
|
|
"epoch": 2.6162440911044262,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004601800865523921,
|
|
"loss": 5.966,
|
|
"mean_token_accuracy": 0.15760403424501418,
|
|
"num_tokens": 5535150.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.0870014190673825,
|
|
"epoch": 2.620541469703481,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00045999103136911204,
|
|
"loss": 5.961,
|
|
"mean_token_accuracy": 0.15921678245067597,
|
|
"num_tokens": 5545121.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.04910888671875,
|
|
"epoch": 2.6248388483025353,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045980157223935965,
|
|
"loss": 5.7994,
|
|
"mean_token_accuracy": 0.16146765872836114,
|
|
"num_tokens": 5553109.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.954406452178955,
|
|
"epoch": 2.62913622690159,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045961170957683806,
|
|
"loss": 5.8505,
|
|
"mean_token_accuracy": 0.167292620241642,
|
|
"num_tokens": 5562020.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.030060815811157,
|
|
"epoch": 2.6334336055006444,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045942144379613147,
|
|
"loss": 5.8499,
|
|
"mean_token_accuracy": 0.15692294687032698,
|
|
"num_tokens": 5571188.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.032728576660157,
|
|
"epoch": 2.637730984099699,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045923077531270426,
|
|
"loss": 5.9149,
|
|
"mean_token_accuracy": 0.1632443591952324,
|
|
"num_tokens": 5580481.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.0685296058654785,
|
|
"epoch": 2.6420283626987535,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004590397045429001,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.16654260009527205,
|
|
"num_tokens": 5589710.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.965172147750854,
|
|
"epoch": 2.646325741297808,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045884823190394134,
|
|
"loss": 5.873,
|
|
"mean_token_accuracy": 0.16391247659921646,
|
|
"num_tokens": 5599083.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.0713605880737305,
|
|
"epoch": 2.650623119896863,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004586563578139275,
|
|
"loss": 5.8052,
|
|
"mean_token_accuracy": 0.16606611907482147,
|
|
"num_tokens": 5607984.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.005927705764771,
|
|
"epoch": 2.6549204984959176,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00045846408269183505,
|
|
"loss": 5.9192,
|
|
"mean_token_accuracy": 0.16373416781425476,
|
|
"num_tokens": 5617470.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.0689976692199705,
|
|
"epoch": 2.659217877094972,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045827140695751603,
|
|
"loss": 5.9445,
|
|
"mean_token_accuracy": 0.15583763122558594,
|
|
"num_tokens": 5626982.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.009213018417358,
|
|
"epoch": 2.6635152556940267,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004580783310316971,
|
|
"loss": 5.8506,
|
|
"mean_token_accuracy": 0.16535259932279586,
|
|
"num_tokens": 5635758.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.069222164154053,
|
|
"epoch": 2.6678126342930812,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045788485533597895,
|
|
"loss": 5.8597,
|
|
"mean_token_accuracy": 0.16713171899318696,
|
|
"num_tokens": 5644890.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.0215541362762455,
|
|
"epoch": 2.672110012892136,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045769098029283526,
|
|
"loss": 5.8504,
|
|
"mean_token_accuracy": 0.1660928949713707,
|
|
"num_tokens": 5653627.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.981487226486206,
|
|
"epoch": 2.6764073914911903,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004574967063256115,
|
|
"loss": 5.8506,
|
|
"mean_token_accuracy": 0.1646118849515915,
|
|
"num_tokens": 5662513.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.006524515151978,
|
|
"epoch": 2.680704770090245,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045730203385852447,
|
|
"loss": 5.9099,
|
|
"mean_token_accuracy": 0.15533654987812043,
|
|
"num_tokens": 5671458.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.062737369537354,
|
|
"epoch": 2.6850021486892994,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000457106963316661,
|
|
"loss": 5.917,
|
|
"mean_token_accuracy": 0.16105935126543044,
|
|
"num_tokens": 5679728.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.054315853118896,
|
|
"epoch": 2.689299527288354,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045691149512597717,
|
|
"loss": 5.9132,
|
|
"mean_token_accuracy": 0.1618688717484474,
|
|
"num_tokens": 5688662.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.997439622879028,
|
|
"epoch": 2.6935969058874085,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045671562971329736,
|
|
"loss": 5.9152,
|
|
"mean_token_accuracy": 0.15352283120155336,
|
|
"num_tokens": 5697442.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.138543939590454,
|
|
"epoch": 2.6978942844864635,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045651936750631337,
|
|
"loss": 5.9403,
|
|
"mean_token_accuracy": 0.154421366751194,
|
|
"num_tokens": 5706496.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.035699129104614,
|
|
"epoch": 2.702191663085518,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045632270893358333,
|
|
"loss": 5.9609,
|
|
"mean_token_accuracy": 0.1593540221452713,
|
|
"num_tokens": 5715954.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.115167951583862,
|
|
"epoch": 2.7064890416845726,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004561256544245312,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.15365591272711754,
|
|
"num_tokens": 5725276.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.042078590393066,
|
|
"epoch": 2.710786420283627,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000455928204409445,
|
|
"loss": 5.9165,
|
|
"mean_token_accuracy": 0.16077258586883544,
|
|
"num_tokens": 5734497.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.959031343460083,
|
|
"epoch": 2.7150837988826817,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045573035931947684,
|
|
"loss": 5.8083,
|
|
"mean_token_accuracy": 0.1680407553911209,
|
|
"num_tokens": 5743539.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.086163187026978,
|
|
"epoch": 2.7193811774817362,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004555321195866411,
|
|
"loss": 5.9085,
|
|
"mean_token_accuracy": 0.16275785565376283,
|
|
"num_tokens": 5752527.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.989049434661865,
|
|
"epoch": 2.723678556080791,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004553334856438143,
|
|
"loss": 5.8398,
|
|
"mean_token_accuracy": 0.16879379898309707,
|
|
"num_tokens": 5761890.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.132557106018067,
|
|
"epoch": 2.7279759346798453,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045513445792473356,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.1575533404946327,
|
|
"num_tokens": 5771041.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.959062623977661,
|
|
"epoch": 2.7322733132789,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004549350368639958,
|
|
"loss": 5.96,
|
|
"mean_token_accuracy": 0.15988690108060838,
|
|
"num_tokens": 5779869.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.043973827362061,
|
|
"epoch": 2.7365706918779544,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00045473522289705693,
|
|
"loss": 5.7947,
|
|
"mean_token_accuracy": 0.1676438882946968,
|
|
"num_tokens": 5788850.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.0489716053009035,
|
|
"epoch": 2.740868070477009,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045453501646023085,
|
|
"loss": 6.0484,
|
|
"mean_token_accuracy": 0.14495702534914018,
|
|
"num_tokens": 5798712.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.087478542327881,
|
|
"epoch": 2.7451654490760635,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045433441799068837,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.15546974316239356,
|
|
"num_tokens": 5808567.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.094131803512573,
|
|
"epoch": 2.749462827675118,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004541334279264562,
|
|
"loss": 5.9076,
|
|
"mean_token_accuracy": 0.16393163800239563,
|
|
"num_tokens": 5817810.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.980616807937622,
|
|
"epoch": 2.7537602062741726,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00045393204670641656,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.15919019430875778,
|
|
"num_tokens": 5827381.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.008398103713989,
|
|
"epoch": 2.758057584873227,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004537302747703055,
|
|
"loss": 5.9211,
|
|
"mean_token_accuracy": 0.16854512244462966,
|
|
"num_tokens": 5836325.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.098405504226685,
|
|
"epoch": 2.7623549634722817,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045352811255871216,
|
|
"loss": 5.8851,
|
|
"mean_token_accuracy": 0.15743474662303925,
|
|
"num_tokens": 5845125.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.998947668075561,
|
|
"epoch": 2.7666523420713363,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045332556051307804,
|
|
"loss": 5.815,
|
|
"mean_token_accuracy": 0.1656831368803978,
|
|
"num_tokens": 5853837.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.0375532627105715,
|
|
"epoch": 2.770949720670391,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045312261907569585,
|
|
"loss": 5.8897,
|
|
"mean_token_accuracy": 0.15993765443563462,
|
|
"num_tokens": 5863832.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.00529670715332,
|
|
"epoch": 2.775247099269446,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00045291928868970867,
|
|
"loss": 5.9357,
|
|
"mean_token_accuracy": 0.15308651626110076,
|
|
"num_tokens": 5872616.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.1932577133178714,
|
|
"epoch": 2.7795444778685003,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004527155697991087,
|
|
"loss": 6.0204,
|
|
"mean_token_accuracy": 0.14987441897392273,
|
|
"num_tokens": 5882153.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.027300119400024,
|
|
"epoch": 2.783841856467555,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004525114628487365,
|
|
"loss": 5.8734,
|
|
"mean_token_accuracy": 0.15730566680431365,
|
|
"num_tokens": 5890650.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.995083427429199,
|
|
"epoch": 2.7881392350666094,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00045230696828428026,
|
|
"loss": 5.8415,
|
|
"mean_token_accuracy": 0.1649170696735382,
|
|
"num_tokens": 5899501.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.9934759616851805,
|
|
"epoch": 2.792436613665664,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004521020865522742,
|
|
"loss": 5.8932,
|
|
"mean_token_accuracy": 0.16182828843593597,
|
|
"num_tokens": 5908040.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.9793510913848875,
|
|
"epoch": 2.7967339922647185,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045189681810009827,
|
|
"loss": 5.7512,
|
|
"mean_token_accuracy": 0.17383025735616683,
|
|
"num_tokens": 5916690.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.033911561965942,
|
|
"epoch": 2.801031370863773,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045169116337597653,
|
|
"loss": 5.8719,
|
|
"mean_token_accuracy": 0.15928059220314025,
|
|
"num_tokens": 5925228.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.977487802505493,
|
|
"epoch": 2.8053287494628276,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000451485122828977,
|
|
"loss": 5.8307,
|
|
"mean_token_accuracy": 0.16225921660661696,
|
|
"num_tokens": 5934289.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.981127738952637,
|
|
"epoch": 2.809626128061882,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045127869690900956,
|
|
"loss": 5.8908,
|
|
"mean_token_accuracy": 0.1589187666773796,
|
|
"num_tokens": 5943516.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.962918758392334,
|
|
"epoch": 2.8139235066609367,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045107188606682613,
|
|
"loss": 5.8196,
|
|
"mean_token_accuracy": 0.16851509660482406,
|
|
"num_tokens": 5951549.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.113703393936158,
|
|
"epoch": 2.8182208852599913,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004508646907540188,
|
|
"loss": 5.9104,
|
|
"mean_token_accuracy": 0.16264692097902297,
|
|
"num_tokens": 5960297.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.091232061386108,
|
|
"epoch": 2.8225182638590463,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004506571114230195,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.15404707491397857,
|
|
"num_tokens": 5969714.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.972896909713745,
|
|
"epoch": 2.826815642458101,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00045044914852709824,
|
|
"loss": 5.8983,
|
|
"mean_token_accuracy": 0.15605547577142714,
|
|
"num_tokens": 5978328.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.044231653213501,
|
|
"epoch": 2.8311130210571553,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004502408025203631,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.15297764241695405,
|
|
"num_tokens": 5987010.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.0108051776885985,
|
|
"epoch": 2.83541039965621,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004500320738577584,
|
|
"loss": 5.8555,
|
|
"mean_token_accuracy": 0.161056350171566,
|
|
"num_tokens": 5996146.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.037520503997802,
|
|
"epoch": 2.8397077782552644,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044982296299506407,
|
|
"loss": 5.9658,
|
|
"mean_token_accuracy": 0.15362147539854049,
|
|
"num_tokens": 6006231.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.06559534072876,
|
|
"epoch": 2.844005156854319,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004496134703888948,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.15250360295176507,
|
|
"num_tokens": 6015795.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.050150346755982,
|
|
"epoch": 2.8483025354533735,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044940359649669846,
|
|
"loss": 5.7708,
|
|
"mean_token_accuracy": 0.17006682828068734,
|
|
"num_tokens": 6024370.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.105326318740845,
|
|
"epoch": 2.852599914052428,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00044919334177675595,
|
|
"loss": 5.9245,
|
|
"mean_token_accuracy": 0.15716814398765563,
|
|
"num_tokens": 6032511.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.938359594345092,
|
|
"epoch": 2.8568972926514826,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044898270668817955,
|
|
"loss": 5.8639,
|
|
"mean_token_accuracy": 0.16547844111919402,
|
|
"num_tokens": 6042039.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.039985847473145,
|
|
"epoch": 2.861194671250537,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000448771691690912,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.15949948728084565,
|
|
"num_tokens": 6051998.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.049960613250732,
|
|
"epoch": 2.8654920498495917,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004485602972457257,
|
|
"loss": 5.8747,
|
|
"mean_token_accuracy": 0.16691060811281205,
|
|
"num_tokens": 6060774.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.023161554336548,
|
|
"epoch": 2.8697894284486463,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044834852381422165,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.163033190369606,
|
|
"num_tokens": 6070915.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.961220645904541,
|
|
"epoch": 2.874086807047701,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044813637185882836,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.15655959397554398,
|
|
"num_tokens": 6080609.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.045425367355347,
|
|
"epoch": 2.8783841856467554,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044792384184280106,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.16156938821077346,
|
|
"num_tokens": 6089231.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.98439588546753,
|
|
"epoch": 2.88268156424581,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044771093423022013,
|
|
"loss": 5.8152,
|
|
"mean_token_accuracy": 0.1664319485425949,
|
|
"num_tokens": 6097835.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.043913793563843,
|
|
"epoch": 2.8869789428448644,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004474976494859909,
|
|
"loss": 5.8943,
|
|
"mean_token_accuracy": 0.16156966090202332,
|
|
"num_tokens": 6107141.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.077665948867798,
|
|
"epoch": 2.891276321443919,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004472839880758419,
|
|
"loss": 5.9429,
|
|
"mean_token_accuracy": 0.1611250728368759,
|
|
"num_tokens": 6116144.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.155654001235962,
|
|
"epoch": 2.8955737000429735,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0004470699504663242,
|
|
"loss": 6.105,
|
|
"mean_token_accuracy": 0.14553092494606973,
|
|
"num_tokens": 6126388.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.927159023284912,
|
|
"epoch": 2.899871078642028,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004468555371248104,
|
|
"loss": 5.8689,
|
|
"mean_token_accuracy": 0.17048726379871368,
|
|
"num_tokens": 6135347.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.08139705657959,
|
|
"epoch": 2.904168457241083,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004466407485194937,
|
|
"loss": 5.9428,
|
|
"mean_token_accuracy": 0.15693512558937073,
|
|
"num_tokens": 6144706.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.054174089431763,
|
|
"epoch": 2.9084658358401376,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004464255851193864,
|
|
"loss": 5.9948,
|
|
"mean_token_accuracy": 0.1642829194664955,
|
|
"num_tokens": 6155417.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.0148406505584715,
|
|
"epoch": 2.912763214439192,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0004462100473943194,
|
|
"loss": 5.7789,
|
|
"mean_token_accuracy": 0.16823015362024307,
|
|
"num_tokens": 6165014.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.02217583656311,
|
|
"epoch": 2.9170605930382467,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000445994135814941,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.16526069045066832,
|
|
"num_tokens": 6174605.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.991291332244873,
|
|
"epoch": 2.9213579716373013,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00044577785085271566,
|
|
"loss": 5.9296,
|
|
"mean_token_accuracy": 0.16107743680477143,
|
|
"num_tokens": 6183419.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.065162229537964,
|
|
"epoch": 2.925655350236356,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004455611929799235,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.16489447951316832,
|
|
"num_tokens": 6193050.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.035895824432373,
|
|
"epoch": 2.9299527288354104,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004453441626696585,
|
|
"loss": 5.8835,
|
|
"mean_token_accuracy": 0.16264651864767074,
|
|
"num_tokens": 6202330.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.982283735275269,
|
|
"epoch": 2.934250107434465,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00044512676039582823,
|
|
"loss": 5.9314,
|
|
"mean_token_accuracy": 0.16290561407804488,
|
|
"num_tokens": 6211195.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.076811933517456,
|
|
"epoch": 2.9385474860335195,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004449089866331524,
|
|
"loss": 5.9219,
|
|
"mean_token_accuracy": 0.1553244933485985,
|
|
"num_tokens": 6221116.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.972243309020996,
|
|
"epoch": 2.942844864632574,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004446908418571617,
|
|
"loss": 5.791,
|
|
"mean_token_accuracy": 0.16958164423704147,
|
|
"num_tokens": 6229270.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.970808219909668,
|
|
"epoch": 2.9471422432316285,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004444723265441973,
|
|
"loss": 5.856,
|
|
"mean_token_accuracy": 0.1619379460811615,
|
|
"num_tokens": 6238687.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.077528619766236,
|
|
"epoch": 2.9514396218306835,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004442534411714092,
|
|
"loss": 5.8469,
|
|
"mean_token_accuracy": 0.1654619261622429,
|
|
"num_tokens": 6247659.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.017802810668945,
|
|
"epoch": 2.955737000429738,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00044403418621675555,
|
|
"loss": 5.922,
|
|
"mean_token_accuracy": 0.16534464359283446,
|
|
"num_tokens": 6256382.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.970675802230835,
|
|
"epoch": 2.9600343790287926,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004438145621590017,
|
|
"loss": 5.8188,
|
|
"mean_token_accuracy": 0.16575789153575898,
|
|
"num_tokens": 6264899.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.018456268310547,
|
|
"epoch": 2.964331757627847,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00044359456947771857,
|
|
"loss": 5.8457,
|
|
"mean_token_accuracy": 0.1677536204457283,
|
|
"num_tokens": 6273557.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.884325647354126,
|
|
"epoch": 2.9686291362269017,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004433742086532824,
|
|
"loss": 5.7924,
|
|
"mean_token_accuracy": 0.17211122661828995,
|
|
"num_tokens": 6283065.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.944413137435913,
|
|
"epoch": 2.9729265148259563,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044315348016687317,
|
|
"loss": 5.8728,
|
|
"mean_token_accuracy": 0.16471027433872223,
|
|
"num_tokens": 6291800.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.050973272323608,
|
|
"epoch": 2.977223893425011,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004429323845004736,
|
|
"loss": 5.9013,
|
|
"mean_token_accuracy": 0.1644984483718872,
|
|
"num_tokens": 6300812.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.981755971908569,
|
|
"epoch": 2.9815212720240654,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044271092213686824,
|
|
"loss": 5.8022,
|
|
"mean_token_accuracy": 0.17195234894752504,
|
|
"num_tokens": 6309407.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.973300933837891,
|
|
"epoch": 2.98581865062312,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00044248909355964247,
|
|
"loss": 5.8329,
|
|
"mean_token_accuracy": 0.16722287237644196,
|
|
"num_tokens": 6317896.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.9300014019012455,
|
|
"epoch": 2.9901160292221745,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044226689925318117,
|
|
"loss": 5.8489,
|
|
"mean_token_accuracy": 0.17007584124803543,
|
|
"num_tokens": 6326492.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.012221193313598,
|
|
"epoch": 2.994413407821229,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044204433970266785,
|
|
"loss": 5.8974,
|
|
"mean_token_accuracy": 0.15638700425624846,
|
|
"num_tokens": 6335552.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.090783786773682,
|
|
"epoch": 2.9987107864202835,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004418214153940837,
|
|
"loss": 5.8862,
|
|
"mean_token_accuracy": 0.16346206068992614,
|
|
"num_tokens": 6344620.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.96839099460178,
|
|
"epoch": 3.002578427159433,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00044159812681420624,
|
|
"loss": 5.6794,
|
|
"mean_token_accuracy": 0.16917475230163997,
|
|
"num_tokens": 6352626.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.934306907653808,
|
|
"epoch": 3.0068758057584875,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004413744744506086,
|
|
"loss": 5.6051,
|
|
"mean_token_accuracy": 0.17078815400600433,
|
|
"num_tokens": 6361618.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.0068758057584875,
|
|
"eval_entropy": 5.8264460177034945,
|
|
"eval_loss": 6.159595489501953,
|
|
"eval_mean_token_accuracy": 0.15812105113254474,
|
|
"eval_num_tokens": 6361618.0,
|
|
"eval_runtime": 2.6438,
|
|
"eval_samples_per_second": 1342.409,
|
|
"eval_steps_per_second": 167.943,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.044166135787964,
|
|
"epoch": 3.011173184357542,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044115045879165806,
|
|
"loss": 5.6149,
|
|
"mean_token_accuracy": 0.1758307546377182,
|
|
"num_tokens": 6370729.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.900899982452392,
|
|
"epoch": 3.0154705629565965,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044092608032651515,
|
|
"loss": 5.512,
|
|
"mean_token_accuracy": 0.18120087534189225,
|
|
"num_tokens": 6380043.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.875496864318848,
|
|
"epoch": 3.019767941555651,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044070133954513305,
|
|
"loss": 5.5029,
|
|
"mean_token_accuracy": 0.1745322197675705,
|
|
"num_tokens": 6389369.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.9651741027832035,
|
|
"epoch": 3.0240653201547056,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004404762369382555,
|
|
"loss": 5.5781,
|
|
"mean_token_accuracy": 0.1736285775899887,
|
|
"num_tokens": 6399619.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.99424991607666,
|
|
"epoch": 3.02836269875376,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00044025077299741683,
|
|
"loss": 5.6368,
|
|
"mean_token_accuracy": 0.16818195283412934,
|
|
"num_tokens": 6408435.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.958137130737304,
|
|
"epoch": 3.0326600773528147,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044002494821494007,
|
|
"loss": 5.6056,
|
|
"mean_token_accuracy": 0.17301858514547347,
|
|
"num_tokens": 6417572.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.98341555595398,
|
|
"epoch": 3.0369574559518693,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043979876308393635,
|
|
"loss": 5.6034,
|
|
"mean_token_accuracy": 0.17133850157260894,
|
|
"num_tokens": 6426153.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.948264408111572,
|
|
"epoch": 3.041254834550924,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004395722180983036,
|
|
"loss": 5.6346,
|
|
"mean_token_accuracy": 0.16839750111103058,
|
|
"num_tokens": 6435258.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.977009344100952,
|
|
"epoch": 3.0455522131499784,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043934531375272535,
|
|
"loss": 5.6944,
|
|
"mean_token_accuracy": 0.1687971070408821,
|
|
"num_tokens": 6445170.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.907476472854614,
|
|
"epoch": 3.049849591749033,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043911805054267015,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.17020042538642882,
|
|
"num_tokens": 6455118.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.960869455337525,
|
|
"epoch": 3.0541469703480875,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00043889042896439004,
|
|
"loss": 5.623,
|
|
"mean_token_accuracy": 0.1676670119166374,
|
|
"num_tokens": 6464168.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.9836524486541744,
|
|
"epoch": 3.0584443489471425,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043866244951491946,
|
|
"loss": 5.6439,
|
|
"mean_token_accuracy": 0.17270424962043762,
|
|
"num_tokens": 6473079.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.978967380523682,
|
|
"epoch": 3.062741727546197,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043843411269207445,
|
|
"loss": 5.6419,
|
|
"mean_token_accuracy": 0.1734544038772583,
|
|
"num_tokens": 6482601.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.911567640304566,
|
|
"epoch": 3.0670391061452515,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004382054189944514,
|
|
"loss": 5.4695,
|
|
"mean_token_accuracy": 0.17878801226615906,
|
|
"num_tokens": 6491089.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.787381601333618,
|
|
"epoch": 3.071336484744306,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004379763689214259,
|
|
"loss": 5.4853,
|
|
"mean_token_accuracy": 0.17799772918224335,
|
|
"num_tokens": 6499484.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.863847255706787,
|
|
"epoch": 3.0756338633433606,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004377469629731518,
|
|
"loss": 5.6207,
|
|
"mean_token_accuracy": 0.17659855186939238,
|
|
"num_tokens": 6508529.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.97723159790039,
|
|
"epoch": 3.079931241942415,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004375172016505599,
|
|
"loss": 5.6394,
|
|
"mean_token_accuracy": 0.17216383069753646,
|
|
"num_tokens": 6517966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.0021164894104,
|
|
"epoch": 3.0842286205414697,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004372870854553572,
|
|
"loss": 5.6152,
|
|
"mean_token_accuracy": 0.16979522556066512,
|
|
"num_tokens": 6527397.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.834334516525269,
|
|
"epoch": 3.0885259991405243,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004370566148900255,
|
|
"loss": 5.5081,
|
|
"mean_token_accuracy": 0.18206476420164108,
|
|
"num_tokens": 6535452.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.909125328063965,
|
|
"epoch": 3.092823377739579,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00043682579045782024,
|
|
"loss": 5.6601,
|
|
"mean_token_accuracy": 0.17319540977478026,
|
|
"num_tokens": 6544591.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.8849709033966064,
|
|
"epoch": 3.0971207563386334,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004365946126627699,
|
|
"loss": 5.5041,
|
|
"mean_token_accuracy": 0.1823827862739563,
|
|
"num_tokens": 6553200.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.894295883178711,
|
|
"epoch": 3.101418134937688,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00043636308200967433,
|
|
"loss": 5.5691,
|
|
"mean_token_accuracy": 0.17766232490539552,
|
|
"num_tokens": 6562632.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.923470449447632,
|
|
"epoch": 3.1057155135367425,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004361311990041039,
|
|
"loss": 5.6058,
|
|
"mean_token_accuracy": 0.16901743412017822,
|
|
"num_tokens": 6571898.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.89887638092041,
|
|
"epoch": 3.110012892135797,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00043589896415239843,
|
|
"loss": 5.6145,
|
|
"mean_token_accuracy": 0.17592914253473282,
|
|
"num_tokens": 6581347.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.930659341812134,
|
|
"epoch": 3.1143102707348516,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043566637796166595,
|
|
"loss": 5.5925,
|
|
"mean_token_accuracy": 0.17468829602003097,
|
|
"num_tokens": 6590148.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.948123455047607,
|
|
"epoch": 3.118607649333906,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.00043543344093978186,
|
|
"loss": 5.6313,
|
|
"mean_token_accuracy": 0.1675386294722557,
|
|
"num_tokens": 6598954.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.928077936172485,
|
|
"epoch": 3.122905027932961,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00043520015359538745,
|
|
"loss": 5.5921,
|
|
"mean_token_accuracy": 0.17654961049556733,
|
|
"num_tokens": 6607787.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.851887369155884,
|
|
"epoch": 3.1272024065320156,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004349665164378891,
|
|
"loss": 5.5785,
|
|
"mean_token_accuracy": 0.17395203113555907,
|
|
"num_tokens": 6617143.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.905450344085693,
|
|
"epoch": 3.13149978513107,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043473252997745684,
|
|
"loss": 5.6188,
|
|
"mean_token_accuracy": 0.16896039694547654,
|
|
"num_tokens": 6625849.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.961460161209106,
|
|
"epoch": 3.1357971637301247,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043449819472502366,
|
|
"loss": 5.6447,
|
|
"mean_token_accuracy": 0.17130472660064697,
|
|
"num_tokens": 6634413.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.886354541778564,
|
|
"epoch": 3.1400945423291793,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004342635111922841,
|
|
"loss": 5.4901,
|
|
"mean_token_accuracy": 0.18334101885557175,
|
|
"num_tokens": 6643785.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.890218591690063,
|
|
"epoch": 3.144391920928234,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004340284798916931,
|
|
"loss": 5.5445,
|
|
"mean_token_accuracy": 0.17555219382047654,
|
|
"num_tokens": 6652802.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.960663986206055,
|
|
"epoch": 3.1486892995272884,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004337931013364653,
|
|
"loss": 5.6401,
|
|
"mean_token_accuracy": 0.17150299698114396,
|
|
"num_tokens": 6662844.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.923622798919678,
|
|
"epoch": 3.152986678126343,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000433557376040573,
|
|
"loss": 5.5504,
|
|
"mean_token_accuracy": 0.17420071959495545,
|
|
"num_tokens": 6671563.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.85622820854187,
|
|
"epoch": 3.1572840567253975,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043332130451874645,
|
|
"loss": 5.5707,
|
|
"mean_token_accuracy": 0.17496092915534972,
|
|
"num_tokens": 6680637.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.967375040054321,
|
|
"epoch": 3.161581435324452,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00043308488728647127,
|
|
"loss": 5.6042,
|
|
"mean_token_accuracy": 0.1741341158747673,
|
|
"num_tokens": 6688892.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.831118774414063,
|
|
"epoch": 3.1658788139235066,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004328481248599882,
|
|
"loss": 5.631,
|
|
"mean_token_accuracy": 0.17212040424346925,
|
|
"num_tokens": 6698376.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.93565878868103,
|
|
"epoch": 3.170176192522561,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004326110177562918,
|
|
"loss": 5.6569,
|
|
"mean_token_accuracy": 0.1700581967830658,
|
|
"num_tokens": 6708279.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.888200664520264,
|
|
"epoch": 3.1744735711216157,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043237356649312926,
|
|
"loss": 5.4974,
|
|
"mean_token_accuracy": 0.1867757335305214,
|
|
"num_tokens": 6716648.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.830375814437867,
|
|
"epoch": 3.17877094972067,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004321357715889991,
|
|
"loss": 5.5559,
|
|
"mean_token_accuracy": 0.17658746093511582,
|
|
"num_tokens": 6725134.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.919095706939697,
|
|
"epoch": 3.1830683283197247,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004318976335631505,
|
|
"loss": 5.5852,
|
|
"mean_token_accuracy": 0.1784835383296013,
|
|
"num_tokens": 6734404.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.9717004776000975,
|
|
"epoch": 3.1873657069187797,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00043165915293558155,
|
|
"loss": 5.6126,
|
|
"mean_token_accuracy": 0.17792525738477707,
|
|
"num_tokens": 6742805.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.871219110488892,
|
|
"epoch": 3.1916630855178343,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004314203302270388,
|
|
"loss": 5.6211,
|
|
"mean_token_accuracy": 0.17564855962991716,
|
|
"num_tokens": 6752178.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.84854793548584,
|
|
"epoch": 3.195960464116889,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004311811659590154,
|
|
"loss": 5.6104,
|
|
"mean_token_accuracy": 0.17023489773273467,
|
|
"num_tokens": 6762018.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.837913227081299,
|
|
"epoch": 3.2002578427159434,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004309416606537507,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.18250211030244828,
|
|
"num_tokens": 6770110.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.831134414672851,
|
|
"epoch": 3.204555221314998,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00043070181483422843,
|
|
"loss": 5.5726,
|
|
"mean_token_accuracy": 0.17923775613307952,
|
|
"num_tokens": 6778364.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.8512858867645265,
|
|
"epoch": 3.2088525999140525,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000430461629024176,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.1771520733833313,
|
|
"num_tokens": 6786860.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.973927116394043,
|
|
"epoch": 3.213149978513107,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004302211037480634,
|
|
"loss": 5.6293,
|
|
"mean_token_accuracy": 0.1735839620232582,
|
|
"num_tokens": 6795983.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.892739915847779,
|
|
"epoch": 3.2174473571121616,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004299802395311015,
|
|
"loss": 5.5765,
|
|
"mean_token_accuracy": 0.17588977962732316,
|
|
"num_tokens": 6804702.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.97205662727356,
|
|
"epoch": 3.221744735711216,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004297390368992414,
|
|
"loss": 5.6232,
|
|
"mean_token_accuracy": 0.17032105773687362,
|
|
"num_tokens": 6813361.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.882761478424072,
|
|
"epoch": 3.2260421143102707,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042949749637917353,
|
|
"loss": 5.6062,
|
|
"mean_token_accuracy": 0.17163125425577164,
|
|
"num_tokens": 6821937.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.841923952102661,
|
|
"epoch": 3.230339492909325,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004292556184983256,
|
|
"loss": 5.6671,
|
|
"mean_token_accuracy": 0.16481948047876357,
|
|
"num_tokens": 6830930.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.894928741455078,
|
|
"epoch": 3.2346368715083798,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004290134037848623,
|
|
"loss": 5.5841,
|
|
"mean_token_accuracy": 0.17584291845560074,
|
|
"num_tokens": 6839815.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.911527681350708,
|
|
"epoch": 3.2389342501074343,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00042877085276768386,
|
|
"loss": 5.6354,
|
|
"mean_token_accuracy": 0.1733119547367096,
|
|
"num_tokens": 6849453.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.83440375328064,
|
|
"epoch": 3.243231628706489,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042852796597642455,
|
|
"loss": 5.5227,
|
|
"mean_token_accuracy": 0.1744684249162674,
|
|
"num_tokens": 6858693.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.843238353729248,
|
|
"epoch": 3.247529007305544,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004282847439414522,
|
|
"loss": 5.5305,
|
|
"mean_token_accuracy": 0.18255263119935988,
|
|
"num_tokens": 6868476.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.898254537582398,
|
|
"epoch": 3.2518263859045984,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004280411871938664,
|
|
"loss": 5.5458,
|
|
"mean_token_accuracy": 0.1825510486960411,
|
|
"num_tokens": 6877471.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.885612916946411,
|
|
"epoch": 3.256123764503653,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004277972962654979,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.17996267527341842,
|
|
"num_tokens": 6886453.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.901696729660034,
|
|
"epoch": 3.2604211431027075,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004275530716889069,
|
|
"loss": 5.637,
|
|
"mean_token_accuracy": 0.17052366137504577,
|
|
"num_tokens": 6895638.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.88498272895813,
|
|
"epoch": 3.264718521701762,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004273085139973822,
|
|
"loss": 5.643,
|
|
"mean_token_accuracy": 0.17624082863330842,
|
|
"num_tokens": 6905127.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.8872064590454105,
|
|
"epoch": 3.2690159003008166,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004270636237249401,
|
|
"loss": 5.5597,
|
|
"mean_token_accuracy": 0.1754902184009552,
|
|
"num_tokens": 6914316.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.850763130187988,
|
|
"epoch": 3.273313278899871,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042681840140632314,
|
|
"loss": 5.6939,
|
|
"mean_token_accuracy": 0.17165953516960145,
|
|
"num_tokens": 6924447.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.896399402618409,
|
|
"epoch": 3.2776106574989257,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004265728475769989,
|
|
"loss": 5.6434,
|
|
"mean_token_accuracy": 0.17049491703510283,
|
|
"num_tokens": 6933279.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.8996035099029545,
|
|
"epoch": 3.28190803609798,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004263269627731586,
|
|
"loss": 5.6562,
|
|
"mean_token_accuracy": 0.17040283232927322,
|
|
"num_tokens": 6941810.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.913550138473511,
|
|
"epoch": 3.2862054146970348,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004260807475317164,
|
|
"loss": 5.6503,
|
|
"mean_token_accuracy": 0.17494209110736847,
|
|
"num_tokens": 6951507.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.847301244735718,
|
|
"epoch": 3.2905027932960893,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004258342023903081,
|
|
"loss": 5.6301,
|
|
"mean_token_accuracy": 0.1691274657845497,
|
|
"num_tokens": 6961047.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.935416173934937,
|
|
"epoch": 3.294800171895144,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00042558732788728975,
|
|
"loss": 5.6031,
|
|
"mean_token_accuracy": 0.17087168544530867,
|
|
"num_tokens": 6969837.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.909591245651245,
|
|
"epoch": 3.2990975504941984,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00042534012456173643,
|
|
"loss": 5.6788,
|
|
"mean_token_accuracy": 0.16828584671020508,
|
|
"num_tokens": 6978581.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.787095212936402,
|
|
"epoch": 3.303394929093253,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042509259295344157,
|
|
"loss": 5.5278,
|
|
"mean_token_accuracy": 0.18176757544279099,
|
|
"num_tokens": 6988929.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.903957509994507,
|
|
"epoch": 3.3076923076923075,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00042484473360291514,
|
|
"loss": 5.5791,
|
|
"mean_token_accuracy": 0.17464792132377624,
|
|
"num_tokens": 6997729.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.907247114181518,
|
|
"epoch": 3.311989686291362,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042459654705138294,
|
|
"loss": 5.6497,
|
|
"mean_token_accuracy": 0.17227049320936202,
|
|
"num_tokens": 7007233.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.83994836807251,
|
|
"epoch": 3.316287064890417,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004243480338407853,
|
|
"loss": 5.5185,
|
|
"mean_token_accuracy": 0.18171052932739257,
|
|
"num_tokens": 7015279.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.885436964035034,
|
|
"epoch": 3.3205844434894716,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004240991945137755,
|
|
"loss": 5.6901,
|
|
"mean_token_accuracy": 0.16825385987758637,
|
|
"num_tokens": 7024579.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.879917812347412,
|
|
"epoch": 3.324881822088526,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00042385002961371944,
|
|
"loss": 5.5882,
|
|
"mean_token_accuracy": 0.18451720774173735,
|
|
"num_tokens": 7033561.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.877239942550659,
|
|
"epoch": 3.3291792006875807,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004236005396846935,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.17509032189846038,
|
|
"num_tokens": 7042695.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.917427015304566,
|
|
"epoch": 3.333476579286635,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00042335072527148406,
|
|
"loss": 5.6244,
|
|
"mean_token_accuracy": 0.17041592448949813,
|
|
"num_tokens": 7051166.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.856033563613892,
|
|
"epoch": 3.3377739578856898,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004231005869195859,
|
|
"loss": 5.5539,
|
|
"mean_token_accuracy": 0.18237031102180482,
|
|
"num_tokens": 7059916.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.913948678970337,
|
|
"epoch": 3.3420713364847443,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004228501251752011,
|
|
"loss": 5.6818,
|
|
"mean_token_accuracy": 0.16955598145723344,
|
|
"num_tokens": 7068813.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.8399125099182125,
|
|
"epoch": 3.346368715083799,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00042259934058523814,
|
|
"loss": 5.6197,
|
|
"mean_token_accuracy": 0.17253002077341079,
|
|
"num_tokens": 7078378.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.839558744430542,
|
|
"epoch": 3.3506660936828534,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00042234823369731027,
|
|
"loss": 5.6707,
|
|
"mean_token_accuracy": 0.17130968123674392,
|
|
"num_tokens": 7088099.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.928657007217407,
|
|
"epoch": 3.354963472281908,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00042209680505973465,
|
|
"loss": 5.6572,
|
|
"mean_token_accuracy": 0.16736788302659988,
|
|
"num_tokens": 7096911.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.80603289604187,
|
|
"epoch": 3.3592608508809625,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004218450552215308,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.1831214651465416,
|
|
"num_tokens": 7106370.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.884301948547363,
|
|
"epoch": 3.363558229480017,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004215929847324199,
|
|
"loss": 5.5957,
|
|
"mean_token_accuracy": 0.17703455686569214,
|
|
"num_tokens": 7115388.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.83881688117981,
|
|
"epoch": 3.3678556080790716,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000421340594142823,
|
|
"loss": 5.5676,
|
|
"mean_token_accuracy": 0.18715633302927018,
|
|
"num_tokens": 7124015.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.908502006530762,
|
|
"epoch": 3.3721529866781266,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042108788400386035,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.17712507843971254,
|
|
"num_tokens": 7132774.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.899597930908203,
|
|
"epoch": 3.376450365277181,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004208348548673498,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.17506725192070008,
|
|
"num_tokens": 7141806.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.956182384490967,
|
|
"epoch": 3.3807477438762357,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000420581507285806,
|
|
"loss": 5.6034,
|
|
"mean_token_accuracy": 0.17655767351388932,
|
|
"num_tokens": 7150570.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.832974672317505,
|
|
"epoch": 3.38504512247529,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004203278418124386,
|
|
"loss": 5.7113,
|
|
"mean_token_accuracy": 0.16485213488340378,
|
|
"num_tokens": 7159811.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.809456825256348,
|
|
"epoch": 3.3893425010743448,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004200738590011518,
|
|
"loss": 5.5489,
|
|
"mean_token_accuracy": 0.17550962716341018,
|
|
"num_tokens": 7169034.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.951978635787964,
|
|
"epoch": 3.3936398796733993,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00041981955940654245,
|
|
"loss": 5.6578,
|
|
"mean_token_accuracy": 0.1656136155128479,
|
|
"num_tokens": 7177384.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.815281343460083,
|
|
"epoch": 3.397937258272454,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004195649435838992,
|
|
"loss": 5.6585,
|
|
"mean_token_accuracy": 0.17446250170469285,
|
|
"num_tokens": 7185774.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.901204252243042,
|
|
"epoch": 3.4022346368715084,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004193100120892013,
|
|
"loss": 5.6254,
|
|
"mean_token_accuracy": 0.17329763919115065,
|
|
"num_tokens": 7195677.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.8513195514678955,
|
|
"epoch": 3.406532015470563,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004190547654791172,
|
|
"loss": 5.6562,
|
|
"mean_token_accuracy": 0.1658914491534233,
|
|
"num_tokens": 7205285.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.855178928375244,
|
|
"epoch": 3.4108293940696175,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00041879920431100347,
|
|
"loss": 5.5171,
|
|
"mean_token_accuracy": 0.18419103920459748,
|
|
"num_tokens": 7213550.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.87856388092041,
|
|
"epoch": 3.415126772668672,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004185433291429036,
|
|
"loss": 5.6757,
|
|
"mean_token_accuracy": 0.1790303811430931,
|
|
"num_tokens": 7222799.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.901208066940308,
|
|
"epoch": 3.4194241512677266,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00041828714053354665,
|
|
"loss": 5.6119,
|
|
"mean_token_accuracy": 0.18016332983970643,
|
|
"num_tokens": 7232041.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.886899948120117,
|
|
"epoch": 3.423721529866781,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004180306390423462,
|
|
"loss": 5.6729,
|
|
"mean_token_accuracy": 0.17059102803468704,
|
|
"num_tokens": 7241236.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.905932283401489,
|
|
"epoch": 3.4280189084658357,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00041777382522939884,
|
|
"loss": 5.6047,
|
|
"mean_token_accuracy": 0.17705892473459245,
|
|
"num_tokens": 7250108.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.910818433761596,
|
|
"epoch": 3.4323162870648902,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00041751669965548344,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.1763069823384285,
|
|
"num_tokens": 7258760.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.829702377319336,
|
|
"epoch": 3.4366136656639448,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00041725926288205945,
|
|
"loss": 5.6843,
|
|
"mean_token_accuracy": 0.17226808667182922,
|
|
"num_tokens": 7267337.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.4366136656639448,
|
|
"eval_entropy": 5.642270052755201,
|
|
"eval_loss": 6.136202335357666,
|
|
"eval_mean_token_accuracy": 0.16109499159092838,
|
|
"eval_num_tokens": 7267337.0,
|
|
"eval_runtime": 2.6535,
|
|
"eval_samples_per_second": 1337.504,
|
|
"eval_steps_per_second": 167.329,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.748719549179077,
|
|
"epoch": 3.4409110442629998,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004170015154712658,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.17859817743301393,
|
|
"num_tokens": 7276640.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.934051370620727,
|
|
"epoch": 3.4452084228620543,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00041674345798591993,
|
|
"loss": 5.6506,
|
|
"mean_token_accuracy": 0.16856029331684114,
|
|
"num_tokens": 7285538.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.840728282928467,
|
|
"epoch": 3.449505801461109,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004164850909895161,
|
|
"loss": 5.5816,
|
|
"mean_token_accuracy": 0.17363331466913223,
|
|
"num_tokens": 7294596.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.810186767578125,
|
|
"epoch": 3.4538031800601634,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004162264150462247,
|
|
"loss": 5.6544,
|
|
"mean_token_accuracy": 0.1704780116677284,
|
|
"num_tokens": 7304463.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.872852516174317,
|
|
"epoch": 3.458100558659218,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00041596743072089065,
|
|
"loss": 5.5591,
|
|
"mean_token_accuracy": 0.18138681650161742,
|
|
"num_tokens": 7314372.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.896686792373657,
|
|
"epoch": 3.4623979372582725,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000415708138579032,
|
|
"loss": 5.5811,
|
|
"mean_token_accuracy": 0.18081890642642975,
|
|
"num_tokens": 7322606.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.864810562133789,
|
|
"epoch": 3.466695315857327,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041544853918683923,
|
|
"loss": 5.6792,
|
|
"mean_token_accuracy": 0.17540037781000137,
|
|
"num_tokens": 7331798.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.799158239364624,
|
|
"epoch": 3.4709926944563816,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004151886331111737,
|
|
"loss": 5.5712,
|
|
"mean_token_accuracy": 0.17659481018781661,
|
|
"num_tokens": 7341182.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.912620210647583,
|
|
"epoch": 3.475290073055436,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041492842091956646,
|
|
"loss": 5.5257,
|
|
"mean_token_accuracy": 0.18442751914262773,
|
|
"num_tokens": 7350419.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.790456485748291,
|
|
"epoch": 3.4795874516544907,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004146679031802167,
|
|
"loss": 5.6379,
|
|
"mean_token_accuracy": 0.16064633280038834,
|
|
"num_tokens": 7359101.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.838149070739746,
|
|
"epoch": 3.4838848302535452,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00041440708046199123,
|
|
"loss": 5.6274,
|
|
"mean_token_accuracy": 0.17170903086662292,
|
|
"num_tokens": 7368376.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.89484167098999,
|
|
"epoch": 3.4881822088525998,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004141459533344226,
|
|
"loss": 5.6038,
|
|
"mean_token_accuracy": 0.17979240119457246,
|
|
"num_tokens": 7377991.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.829446649551391,
|
|
"epoch": 3.4924795874516543,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041388452236770795,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.17151045948266982,
|
|
"num_tokens": 7387712.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.854921340942383,
|
|
"epoch": 3.4967769660507093,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00041362278813270823,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.18058791309595107,
|
|
"num_tokens": 7396151.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.914624977111816,
|
|
"epoch": 3.501074344649764,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00041336075120094616,
|
|
"loss": 5.671,
|
|
"mean_token_accuracy": 0.16610753834247588,
|
|
"num_tokens": 7405090.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.888888692855835,
|
|
"epoch": 3.5053717232488184,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00041309841214460586,
|
|
"loss": 5.5984,
|
|
"mean_token_accuracy": 0.18103182762861253,
|
|
"num_tokens": 7413243.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.767253637313843,
|
|
"epoch": 3.509669101847873,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004128357715365309,
|
|
"loss": 5.5845,
|
|
"mean_token_accuracy": 0.17408550381660462,
|
|
"num_tokens": 7421382.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.866237306594849,
|
|
"epoch": 3.5139664804469275,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041257282995022345,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.16313876658678056,
|
|
"num_tokens": 7431401.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.808319807052612,
|
|
"epoch": 3.518263859045982,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004123095879598426,
|
|
"loss": 5.5712,
|
|
"mean_token_accuracy": 0.17768578231334686,
|
|
"num_tokens": 7441092.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.76686840057373,
|
|
"epoch": 3.5225612376450366,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00041204604614020397,
|
|
"loss": 5.636,
|
|
"mean_token_accuracy": 0.17856015861034394,
|
|
"num_tokens": 7450558.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.9073854923248295,
|
|
"epoch": 3.526858616244091,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.0004117822050667773,
|
|
"loss": 5.6369,
|
|
"mean_token_accuracy": 0.17222602814435958,
|
|
"num_tokens": 7459622.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.9487738609313965,
|
|
"epoch": 3.5311559948431457,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00041151806531568617,
|
|
"loss": 5.5886,
|
|
"mean_token_accuracy": 0.17527564316987992,
|
|
"num_tokens": 7468311.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.880669832229614,
|
|
"epoch": 3.5354533734422002,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00041125362746370625,
|
|
"loss": 5.6243,
|
|
"mean_token_accuracy": 0.16762883216142654,
|
|
"num_tokens": 7477049.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.83824610710144,
|
|
"epoch": 3.5397507520412548,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004109888920882639,
|
|
"loss": 5.6271,
|
|
"mean_token_accuracy": 0.1759859338402748,
|
|
"num_tokens": 7486332.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.865653371810913,
|
|
"epoch": 3.5440481306403093,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004107238597674356,
|
|
"loss": 5.5881,
|
|
"mean_token_accuracy": 0.17611861675977708,
|
|
"num_tokens": 7494713.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.840641164779663,
|
|
"epoch": 3.548345509239364,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000410458531079946,
|
|
"loss": 5.5783,
|
|
"mean_token_accuracy": 0.17697491496801376,
|
|
"num_tokens": 7504359.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.859061908721924,
|
|
"epoch": 3.5526428878384184,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004101929066051668,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.18038901388645173,
|
|
"num_tokens": 7512792.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.813077354431153,
|
|
"epoch": 3.556940266437473,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 0.0004099269869231157,
|
|
"loss": 5.5723,
|
|
"mean_token_accuracy": 0.17631174772977828,
|
|
"num_tokens": 7522617.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.7907510757446286,
|
|
"epoch": 3.5612376450365275,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00040966077261445495,
|
|
"loss": 5.6126,
|
|
"mean_token_accuracy": 0.1791912406682968,
|
|
"num_tokens": 7531123.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.863047933578491,
|
|
"epoch": 3.565535023635582,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004093942642604904,
|
|
"loss": 5.6927,
|
|
"mean_token_accuracy": 0.16946851313114167,
|
|
"num_tokens": 7541521.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.871539163589477,
|
|
"epoch": 3.5698324022346366,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00040912746244316944,
|
|
"loss": 5.5418,
|
|
"mean_token_accuracy": 0.1836775541305542,
|
|
"num_tokens": 7550343.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.896195697784424,
|
|
"epoch": 3.5741297808336916,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00040886036774508095,
|
|
"loss": 5.7998,
|
|
"mean_token_accuracy": 0.15613570734858512,
|
|
"num_tokens": 7560399.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.9076111793518065,
|
|
"epoch": 3.578427159432746,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004085929807494527,
|
|
"loss": 5.5637,
|
|
"mean_token_accuracy": 0.17794504016637802,
|
|
"num_tokens": 7569982.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.705628967285156,
|
|
"epoch": 3.5827245380318007,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004083253020401512,
|
|
"loss": 5.4864,
|
|
"mean_token_accuracy": 0.18662448078393937,
|
|
"num_tokens": 7578954.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.850636005401611,
|
|
"epoch": 3.5870219166308552,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004080573322016797,
|
|
"loss": 5.5914,
|
|
"mean_token_accuracy": 0.17876261919736863,
|
|
"num_tokens": 7587424.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.819705867767334,
|
|
"epoch": 3.59131929522991,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004077890718191773,
|
|
"loss": 5.4938,
|
|
"mean_token_accuracy": 0.18184783905744553,
|
|
"num_tokens": 7596900.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.852156782150269,
|
|
"epoch": 3.5956166738289643,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00040752052147841733,
|
|
"loss": 5.708,
|
|
"mean_token_accuracy": 0.17140453606843947,
|
|
"num_tokens": 7606774.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.848931837081909,
|
|
"epoch": 3.599914052428019,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004072516817658065,
|
|
"loss": 5.6247,
|
|
"mean_token_accuracy": 0.17357547879219054,
|
|
"num_tokens": 7615691.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.918767261505127,
|
|
"epoch": 3.6042114310270734,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0004069825532683831,
|
|
"loss": 5.6092,
|
|
"mean_token_accuracy": 0.18187319487333298,
|
|
"num_tokens": 7625294.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.816747379302979,
|
|
"epoch": 3.608508809626128,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00040671313657381645,
|
|
"loss": 5.6108,
|
|
"mean_token_accuracy": 0.17615855038166045,
|
|
"num_tokens": 7634563.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.858688306808472,
|
|
"epoch": 3.6128061882251825,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040644343227040473,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.17691098898649216,
|
|
"num_tokens": 7643860.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.874885368347168,
|
|
"epoch": 3.617103566824237,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004061734409470745,
|
|
"loss": 5.6585,
|
|
"mean_token_accuracy": 0.16745558083057405,
|
|
"num_tokens": 7654142.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.879082441329956,
|
|
"epoch": 3.621400945423292,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004059031631933788,
|
|
"loss": 5.6233,
|
|
"mean_token_accuracy": 0.17368015199899672,
|
|
"num_tokens": 7662910.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.85358247756958,
|
|
"epoch": 3.6256983240223466,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00040563259959949615,
|
|
"loss": 5.584,
|
|
"mean_token_accuracy": 0.17502041161060333,
|
|
"num_tokens": 7670764.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.916553735733032,
|
|
"epoch": 3.629995702621401,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004053617507562295,
|
|
"loss": 5.6722,
|
|
"mean_token_accuracy": 0.16264282315969467,
|
|
"num_tokens": 7679422.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.863842535018921,
|
|
"epoch": 3.6342930812204557,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00040509061725500426,
|
|
"loss": 5.5811,
|
|
"mean_token_accuracy": 0.17766340523958207,
|
|
"num_tokens": 7687992.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.883695793151856,
|
|
"epoch": 3.6385904598195102,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004048191996878677,
|
|
"loss": 5.6676,
|
|
"mean_token_accuracy": 0.1716696083545685,
|
|
"num_tokens": 7696607.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.971418380737305,
|
|
"epoch": 3.642887838418565,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00040454749864748734,
|
|
"loss": 5.7029,
|
|
"mean_token_accuracy": 0.16839040964841842,
|
|
"num_tokens": 7705280.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.892542886734009,
|
|
"epoch": 3.6471852170176193,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004042755147271496,
|
|
"loss": 5.5647,
|
|
"mean_token_accuracy": 0.17813005894422532,
|
|
"num_tokens": 7714120.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.797539710998535,
|
|
"epoch": 3.651482595616674,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004040032485207587,
|
|
"loss": 5.6175,
|
|
"mean_token_accuracy": 0.17526279091835023,
|
|
"num_tokens": 7724627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.817934894561768,
|
|
"epoch": 3.6557799742157284,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004037307006228352,
|
|
"loss": 5.6781,
|
|
"mean_token_accuracy": 0.16682509928941727,
|
|
"num_tokens": 7734650.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.882663631439209,
|
|
"epoch": 3.660077352814783,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004034578716285147,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.18512446582317352,
|
|
"num_tokens": 7744357.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.914453601837158,
|
|
"epoch": 3.6643747314138375,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004031847621335467,
|
|
"loss": 5.5738,
|
|
"mean_token_accuracy": 0.16555424332618712,
|
|
"num_tokens": 7753207.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.882584476470948,
|
|
"epoch": 3.668672110012892,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004029113727342933,
|
|
"loss": 5.6791,
|
|
"mean_token_accuracy": 0.1736137792468071,
|
|
"num_tokens": 7762750.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.891817855834961,
|
|
"epoch": 3.6729694886119466,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040263770402772746,
|
|
"loss": 5.5961,
|
|
"mean_token_accuracy": 0.17807978689670562,
|
|
"num_tokens": 7771539.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.8590913772583,
|
|
"epoch": 3.677266867211001,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004023637566114325,
|
|
"loss": 5.6903,
|
|
"mean_token_accuracy": 0.1734776258468628,
|
|
"num_tokens": 7781447.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.854595947265625,
|
|
"epoch": 3.6815642458100557,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004020895310835999,
|
|
"loss": 5.5859,
|
|
"mean_token_accuracy": 0.1780365601181984,
|
|
"num_tokens": 7789977.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.943436479568481,
|
|
"epoch": 3.6858616244091102,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00040181502804302865,
|
|
"loss": 5.7167,
|
|
"mean_token_accuracy": 0.1653963655233383,
|
|
"num_tokens": 7800000.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.882033729553223,
|
|
"epoch": 3.690159003008165,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00040154024808912377,
|
|
"loss": 5.7234,
|
|
"mean_token_accuracy": 0.16748702079057692,
|
|
"num_tokens": 7809062.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.827185010910034,
|
|
"epoch": 3.6944563816072193,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004012651918218947,
|
|
"loss": 5.5224,
|
|
"mean_token_accuracy": 0.1866112694144249,
|
|
"num_tokens": 7817136.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.838236951828003,
|
|
"epoch": 3.6987537602062743,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004009898598419544,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.17415323108434677,
|
|
"num_tokens": 7825278.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.827865552902222,
|
|
"epoch": 3.703051138805329,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000400714252750518,
|
|
"loss": 5.6588,
|
|
"mean_token_accuracy": 0.1713027611374855,
|
|
"num_tokens": 7833448.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.852033996582032,
|
|
"epoch": 3.7073485174043834,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004004383711494011,
|
|
"loss": 5.5766,
|
|
"mean_token_accuracy": 0.17906928509473802,
|
|
"num_tokens": 7841919.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.922611665725708,
|
|
"epoch": 3.711645896003438,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004001622156410189,
|
|
"loss": 5.7384,
|
|
"mean_token_accuracy": 0.16610016524791718,
|
|
"num_tokens": 7851584.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.922744607925415,
|
|
"epoch": 3.7159432746024925,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00039988578682838467,
|
|
"loss": 5.5952,
|
|
"mean_token_accuracy": 0.18116231262683868,
|
|
"num_tokens": 7860329.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.847246217727661,
|
|
"epoch": 3.720240653201547,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00039960908531510843,
|
|
"loss": 5.7093,
|
|
"mean_token_accuracy": 0.17550332099199295,
|
|
"num_tokens": 7870393.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.8290520191192625,
|
|
"epoch": 3.7245380318006016,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003993321117053956,
|
|
"loss": 5.6287,
|
|
"mean_token_accuracy": 0.1815280094742775,
|
|
"num_tokens": 7878776.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.988626194000244,
|
|
"epoch": 3.728835410399656,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00039905486660404604,
|
|
"loss": 5.724,
|
|
"mean_token_accuracy": 0.16447362005710603,
|
|
"num_tokens": 7888232.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.88866777420044,
|
|
"epoch": 3.7331327889987107,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00039877735061645206,
|
|
"loss": 5.6719,
|
|
"mean_token_accuracy": 0.16899960935115815,
|
|
"num_tokens": 7898951.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.883985328674316,
|
|
"epoch": 3.7374301675977653,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003984995643485977,
|
|
"loss": 5.7271,
|
|
"mean_token_accuracy": 0.1719050094485283,
|
|
"num_tokens": 7908162.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.8497092723846436,
|
|
"epoch": 3.74172754619682,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00039822150840705716,
|
|
"loss": 5.5964,
|
|
"mean_token_accuracy": 0.17865013629198073,
|
|
"num_tokens": 7916953.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.873455142974853,
|
|
"epoch": 3.746024924795875,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00039794318339899347,
|
|
"loss": 5.6154,
|
|
"mean_token_accuracy": 0.17540279328823088,
|
|
"num_tokens": 7925622.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.882204151153564,
|
|
"epoch": 3.7503223033949293,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039766458993215726,
|
|
"loss": 5.6714,
|
|
"mean_token_accuracy": 0.17278452664613725,
|
|
"num_tokens": 7934807.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.854500770568848,
|
|
"epoch": 3.754619681993984,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039738572861488527,
|
|
"loss": 5.6957,
|
|
"mean_token_accuracy": 0.16906797662377357,
|
|
"num_tokens": 7945864.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.845993137359619,
|
|
"epoch": 3.7589170605930384,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000397106600056099,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.18473466038703917,
|
|
"num_tokens": 7954571.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.850477170944214,
|
|
"epoch": 3.763214439192093,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003968272048653039,
|
|
"loss": 5.5346,
|
|
"mean_token_accuracy": 0.18714721649885177,
|
|
"num_tokens": 7963092.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.799531698226929,
|
|
"epoch": 3.7675118177911475,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003965475436525873,
|
|
"loss": 5.6352,
|
|
"mean_token_accuracy": 0.1693410098552704,
|
|
"num_tokens": 7972509.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.848278999328613,
|
|
"epoch": 3.771809196390202,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003962676170286174,
|
|
"loss": 5.5674,
|
|
"mean_token_accuracy": 0.18184708207845687,
|
|
"num_tokens": 7981494.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.827572202682495,
|
|
"epoch": 3.7761065749892566,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039598742560464223,
|
|
"loss": 5.5835,
|
|
"mean_token_accuracy": 0.17856394052505492,
|
|
"num_tokens": 7991788.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.707248544692993,
|
|
"epoch": 3.780403953588311,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003957069699924877,
|
|
"loss": 5.5244,
|
|
"mean_token_accuracy": 0.18229787796735764,
|
|
"num_tokens": 8000528.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.847303819656372,
|
|
"epoch": 3.7847013321873657,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000395426250804557,
|
|
"loss": 5.6155,
|
|
"mean_token_accuracy": 0.174784092605114,
|
|
"num_tokens": 8008729.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.894458198547364,
|
|
"epoch": 3.7889987107864203,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039514526865382847,
|
|
"loss": 5.6985,
|
|
"mean_token_accuracy": 0.16691654771566392,
|
|
"num_tokens": 8017391.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.902827262878418,
|
|
"epoch": 3.793296089385475,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003948640241538548,
|
|
"loss": 5.6189,
|
|
"mean_token_accuracy": 0.17787647843360901,
|
|
"num_tokens": 8026772.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.896196174621582,
|
|
"epoch": 3.7975934679845293,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003945825179187617,
|
|
"loss": 5.7145,
|
|
"mean_token_accuracy": 0.16603004932403564,
|
|
"num_tokens": 8035231.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.851997947692871,
|
|
"epoch": 3.801890846583584,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039430075056324604,
|
|
"loss": 5.6748,
|
|
"mean_token_accuracy": 0.16944888830184937,
|
|
"num_tokens": 8044097.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.841471242904663,
|
|
"epoch": 3.8061882251826384,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00039401872270257546,
|
|
"loss": 5.6385,
|
|
"mean_token_accuracy": 0.1733471557497978,
|
|
"num_tokens": 8053511.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.968043851852417,
|
|
"epoch": 3.810485603781693,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039373643495258567,
|
|
"loss": 5.7047,
|
|
"mean_token_accuracy": 0.17462301403284072,
|
|
"num_tokens": 8063237.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.77347617149353,
|
|
"epoch": 3.8147829823807475,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039345388792968056,
|
|
"loss": 5.5551,
|
|
"mean_token_accuracy": 0.1744810625910759,
|
|
"num_tokens": 8072098.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.869391584396363,
|
|
"epoch": 3.819080360979802,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00039317108225082984,
|
|
"loss": 5.663,
|
|
"mean_token_accuracy": 0.16630891263484954,
|
|
"num_tokens": 8080789.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.875336980819702,
|
|
"epoch": 3.8233777395788566,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039288801853356806,
|
|
"loss": 5.658,
|
|
"mean_token_accuracy": 0.1675410211086273,
|
|
"num_tokens": 8091013.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.8768517017364506,
|
|
"epoch": 3.8276751181779116,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003926046973959932,
|
|
"loss": 5.5909,
|
|
"mean_token_accuracy": 0.17918838262557985,
|
|
"num_tokens": 8099857.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.836393642425537,
|
|
"epoch": 3.831972496776966,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003923211194567654,
|
|
"loss": 5.5621,
|
|
"mean_token_accuracy": 0.176430743932724,
|
|
"num_tokens": 8109268.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.873169946670532,
|
|
"epoch": 3.8362698753760207,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039203728533510556,
|
|
"loss": 5.6671,
|
|
"mean_token_accuracy": 0.16467680782079697,
|
|
"num_tokens": 8119242.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.933674383163452,
|
|
"epoch": 3.8405672539750753,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000391753195650794,
|
|
"loss": 5.6683,
|
|
"mean_token_accuracy": 0.1716270163655281,
|
|
"num_tokens": 8129031.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.863985443115235,
|
|
"epoch": 3.84486463257413,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00039146885102416895,
|
|
"loss": 5.6356,
|
|
"mean_token_accuracy": 0.17117024064064026,
|
|
"num_tokens": 8137550.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.883577060699463,
|
|
"epoch": 3.8491620111731844,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039118425207612553,
|
|
"loss": 5.6485,
|
|
"mean_token_accuracy": 0.17631032764911653,
|
|
"num_tokens": 8146410.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.856779003143311,
|
|
"epoch": 3.853459389772239,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039089939942811396,
|
|
"loss": 5.5395,
|
|
"mean_token_accuracy": 0.17506932467222214,
|
|
"num_tokens": 8155126.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.715206813812256,
|
|
"epoch": 3.8577567683712934,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00039061429370213863,
|
|
"loss": 5.5223,
|
|
"mean_token_accuracy": 0.18575454652309417,
|
|
"num_tokens": 8163623.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.765589618682862,
|
|
"epoch": 3.862054146970348,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00039032893552075646,
|
|
"loss": 5.5837,
|
|
"mean_token_accuracy": 0.17770327031612396,
|
|
"num_tokens": 8172446.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.868394041061402,
|
|
"epoch": 3.8663515255694025,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003900433255070758,
|
|
"loss": 5.6957,
|
|
"mean_token_accuracy": 0.17402739822864532,
|
|
"num_tokens": 8181278.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.8663515255694025,
|
|
"eval_entropy": 5.843679216530946,
|
|
"eval_loss": 6.088552474975586,
|
|
"eval_mean_token_accuracy": 0.16329166266295286,
|
|
"eval_num_tokens": 8181278.0,
|
|
"eval_runtime": 2.643,
|
|
"eval_samples_per_second": 1342.786,
|
|
"eval_steps_per_second": 167.99,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.937895774841309,
|
|
"epoch": 3.870648904168457,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00038975746428475454,
|
|
"loss": 5.6742,
|
|
"mean_token_accuracy": 0.17236708998680114,
|
|
"num_tokens": 8189981.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.849340963363647,
|
|
"epoch": 3.874946282767512,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00038947135247799955,
|
|
"loss": 5.6665,
|
|
"mean_token_accuracy": 0.1750643029808998,
|
|
"num_tokens": 8198674.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.797895860671997,
|
|
"epoch": 3.8792436613665666,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00038918499071156443,
|
|
"loss": 5.6117,
|
|
"mean_token_accuracy": 0.17492095679044722,
|
|
"num_tokens": 8208012.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.86012601852417,
|
|
"epoch": 3.883541039965621,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000388898379610749,
|
|
"loss": 5.5476,
|
|
"mean_token_accuracy": 0.17556158900260926,
|
|
"num_tokens": 8216811.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.908657836914062,
|
|
"epoch": 3.8878384185646757,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003886115198013973,
|
|
"loss": 5.7455,
|
|
"mean_token_accuracy": 0.16469820439815522,
|
|
"num_tokens": 8226637.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.914604520797729,
|
|
"epoch": 3.8921357971637303,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003883244119098965,
|
|
"loss": 5.6479,
|
|
"mean_token_accuracy": 0.1728173092007637,
|
|
"num_tokens": 8235443.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.91871485710144,
|
|
"epoch": 3.896433175762785,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003880370565631754,
|
|
"loss": 5.653,
|
|
"mean_token_accuracy": 0.16667632162570953,
|
|
"num_tokens": 8243791.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.868415021896363,
|
|
"epoch": 3.9007305543618394,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00038774945438870337,
|
|
"loss": 5.6173,
|
|
"mean_token_accuracy": 0.16614460945129395,
|
|
"num_tokens": 8252831.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.831996297836303,
|
|
"epoch": 3.905027932960894,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038746160601448845,
|
|
"loss": 5.6458,
|
|
"mean_token_accuracy": 0.16777586340904235,
|
|
"num_tokens": 8262281.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.863851070404053,
|
|
"epoch": 3.9093253115599484,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003871735120690766,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.18136228770017623,
|
|
"num_tokens": 8270834.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.855620241165161,
|
|
"epoch": 3.913622690159003,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003868851731815497,
|
|
"loss": 5.6159,
|
|
"mean_token_accuracy": 0.1795090541243553,
|
|
"num_tokens": 8280674.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.789209604263306,
|
|
"epoch": 3.9179200687580575,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003865965899815247,
|
|
"loss": 5.6152,
|
|
"mean_token_accuracy": 0.1722850039601326,
|
|
"num_tokens": 8289662.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.864688396453857,
|
|
"epoch": 3.922217447357112,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003863077630991518,
|
|
"loss": 5.5681,
|
|
"mean_token_accuracy": 0.17464204430580138,
|
|
"num_tokens": 8298795.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.8732294082641605,
|
|
"epoch": 3.9265148259561666,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003860186931651139,
|
|
"loss": 5.6042,
|
|
"mean_token_accuracy": 0.17186138331890105,
|
|
"num_tokens": 8306855.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.867685842514038,
|
|
"epoch": 3.930812204555221,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003857293808106238,
|
|
"loss": 5.6473,
|
|
"mean_token_accuracy": 0.17051120847463608,
|
|
"num_tokens": 8316041.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.818361806869507,
|
|
"epoch": 3.9351095831542757,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003854398266674241,
|
|
"loss": 5.623,
|
|
"mean_token_accuracy": 0.17779334932565688,
|
|
"num_tokens": 8325873.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.825986337661743,
|
|
"epoch": 3.9394069617533303,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00038515003136778544,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.1777907833456993,
|
|
"num_tokens": 8334786.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.847596263885498,
|
|
"epoch": 3.943704340352385,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038485999554450483,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.1803155705332756,
|
|
"num_tokens": 8345189.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.77664589881897,
|
|
"epoch": 3.9480017189514394,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00038456971983090454,
|
|
"loss": 5.6015,
|
|
"mean_token_accuracy": 0.17905200719833375,
|
|
"num_tokens": 8353537.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.848537635803223,
|
|
"epoch": 3.9522990975504944,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003842792048608309,
|
|
"loss": 5.6205,
|
|
"mean_token_accuracy": 0.17351499795913697,
|
|
"num_tokens": 8363477.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.818036937713623,
|
|
"epoch": 3.956596476149549,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003839884512686523,
|
|
"loss": 5.6374,
|
|
"mean_token_accuracy": 0.17440737187862396,
|
|
"num_tokens": 8372501.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.823426055908203,
|
|
"epoch": 3.9608938547486034,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00038369745968925846,
|
|
"loss": 5.6355,
|
|
"mean_token_accuracy": 0.17501891255378724,
|
|
"num_tokens": 8381457.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.855378866195679,
|
|
"epoch": 3.965191233347658,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00038340623075805875,
|
|
"loss": 5.6789,
|
|
"mean_token_accuracy": 0.16702238023281096,
|
|
"num_tokens": 8390506.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.886463594436646,
|
|
"epoch": 3.9694886119467125,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038311476511098053,
|
|
"loss": 5.7195,
|
|
"mean_token_accuracy": 0.167725171148777,
|
|
"num_tokens": 8399803.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.836212968826294,
|
|
"epoch": 3.973785990545767,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003828230633844685,
|
|
"loss": 5.5963,
|
|
"mean_token_accuracy": 0.1765116944909096,
|
|
"num_tokens": 8408531.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.745146036148071,
|
|
"epoch": 3.9780833691448216,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00038253112621548243,
|
|
"loss": 5.5118,
|
|
"mean_token_accuracy": 0.18591788560152053,
|
|
"num_tokens": 8416721.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.8053804397583,
|
|
"epoch": 3.982380747743876,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003822389542414966,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.18439393490552902,
|
|
"num_tokens": 8425896.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.862294912338257,
|
|
"epoch": 3.9866781263429307,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00038194654810049775,
|
|
"loss": 5.6013,
|
|
"mean_token_accuracy": 0.17082200497388839,
|
|
"num_tokens": 8434695.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.85670313835144,
|
|
"epoch": 3.9909755049419853,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000381653908430984,
|
|
"loss": 5.7118,
|
|
"mean_token_accuracy": 0.164557521045208,
|
|
"num_tokens": 8443696.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.902082681655884,
|
|
"epoch": 3.99527288354104,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003813610358719634,
|
|
"loss": 5.6086,
|
|
"mean_token_accuracy": 0.17765663415193558,
|
|
"num_tokens": 8453429.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.891878366470337,
|
|
"epoch": 3.999570262140095,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00038106793106295266,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.1712706670165062,
|
|
"num_tokens": 8462823.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.836284690433079,
|
|
"epoch": 4.003437902879243,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003807745946439754,
|
|
"loss": 5.5523,
|
|
"mean_token_accuracy": 0.18386898934841156,
|
|
"num_tokens": 8471681.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.849593925476074,
|
|
"epoch": 4.007735281478298,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003804810272555612,
|
|
"loss": 5.3618,
|
|
"mean_token_accuracy": 0.18703516721725463,
|
|
"num_tokens": 8480819.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.744481754302979,
|
|
"epoch": 4.012032660077352,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003801872295387439,
|
|
"loss": 5.3175,
|
|
"mean_token_accuracy": 0.1954621285200119,
|
|
"num_tokens": 8489692.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.813962602615357,
|
|
"epoch": 4.016330038676408,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003798932021350603,
|
|
"loss": 5.4014,
|
|
"mean_token_accuracy": 0.18128430843353271,
|
|
"num_tokens": 8498523.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.84787540435791,
|
|
"epoch": 4.020627417275462,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00037959894568654864,
|
|
"loss": 5.4201,
|
|
"mean_token_accuracy": 0.1808469995856285,
|
|
"num_tokens": 8507641.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.777582311630249,
|
|
"epoch": 4.024924795874517,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003793044608357474,
|
|
"loss": 5.2968,
|
|
"mean_token_accuracy": 0.1901951923966408,
|
|
"num_tokens": 8516082.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.872768354415894,
|
|
"epoch": 4.0292221744735714,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003790097482256939,
|
|
"loss": 5.4549,
|
|
"mean_token_accuracy": 0.18143108785152434,
|
|
"num_tokens": 8526424.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.858446884155273,
|
|
"epoch": 4.033519553072626,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003787148084999225,
|
|
"loss": 5.3228,
|
|
"mean_token_accuracy": 0.18997532576322557,
|
|
"num_tokens": 8536729.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.792700862884521,
|
|
"epoch": 4.0378169316716805,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00037841964230246394,
|
|
"loss": 5.4003,
|
|
"mean_token_accuracy": 0.17986347377300263,
|
|
"num_tokens": 8546250.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.789514112472534,
|
|
"epoch": 4.042114310270735,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003781242502778429,
|
|
"loss": 5.3779,
|
|
"mean_token_accuracy": 0.19723970144987107,
|
|
"num_tokens": 8555067.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.784922695159912,
|
|
"epoch": 4.04641168886979,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00037782863307107785,
|
|
"loss": 5.3141,
|
|
"mean_token_accuracy": 0.19035983383655547,
|
|
"num_tokens": 8563638.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.759968709945679,
|
|
"epoch": 4.050709067468844,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00037753279132767833,
|
|
"loss": 5.4019,
|
|
"mean_token_accuracy": 0.18232161700725555,
|
|
"num_tokens": 8572233.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.7771971225738525,
|
|
"epoch": 4.055006446067899,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00037723672569364453,
|
|
"loss": 5.3576,
|
|
"mean_token_accuracy": 0.1858087807893753,
|
|
"num_tokens": 8580995.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.8218186378479,
|
|
"epoch": 4.059303824666953,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00037694043681546545,
|
|
"loss": 5.3341,
|
|
"mean_token_accuracy": 0.18529362380504608,
|
|
"num_tokens": 8590250.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.74794716835022,
|
|
"epoch": 4.063601203266008,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003766439253401177,
|
|
"loss": 5.3648,
|
|
"mean_token_accuracy": 0.18297045677900314,
|
|
"num_tokens": 8599183.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.743490505218506,
|
|
"epoch": 4.067898581865062,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00037634719191506367,
|
|
"loss": 5.3254,
|
|
"mean_token_accuracy": 0.19299045205116272,
|
|
"num_tokens": 8607085.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.857343959808349,
|
|
"epoch": 4.072195960464117,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00037605023718825065,
|
|
"loss": 5.3829,
|
|
"mean_token_accuracy": 0.18608618527650833,
|
|
"num_tokens": 8616308.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.866016149520874,
|
|
"epoch": 4.0764933390631715,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000375753061808109,
|
|
"loss": 5.4497,
|
|
"mean_token_accuracy": 0.18775359839200972,
|
|
"num_tokens": 8625337.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.824159097671509,
|
|
"epoch": 4.080790717662226,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00037545566642355107,
|
|
"loss": 5.4219,
|
|
"mean_token_accuracy": 0.18650178164243697,
|
|
"num_tokens": 8634389.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.84695987701416,
|
|
"epoch": 4.0850880962612806,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003751580516839695,
|
|
"loss": 5.518,
|
|
"mean_token_accuracy": 0.1780924454331398,
|
|
"num_tokens": 8644691.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.765744876861572,
|
|
"epoch": 4.089385474860335,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00037486021823923574,
|
|
"loss": 5.426,
|
|
"mean_token_accuracy": 0.1831970378756523,
|
|
"num_tokens": 8653775.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.841619539260864,
|
|
"epoch": 4.09368285345939,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00037456216673969925,
|
|
"loss": 5.458,
|
|
"mean_token_accuracy": 0.18304252177476882,
|
|
"num_tokens": 8661958.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.88407793045044,
|
|
"epoch": 4.097980232058444,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003742638978361851,
|
|
"loss": 5.3529,
|
|
"mean_token_accuracy": 0.19176291823387145,
|
|
"num_tokens": 8671432.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.699429941177368,
|
|
"epoch": 4.102277610657499,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00037396541217999367,
|
|
"loss": 5.3526,
|
|
"mean_token_accuracy": 0.19269220978021623,
|
|
"num_tokens": 8681778.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.763651275634766,
|
|
"epoch": 4.106574989256553,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003736667104228981,
|
|
"loss": 5.4269,
|
|
"mean_token_accuracy": 0.18959370255470276,
|
|
"num_tokens": 8690769.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.855369663238525,
|
|
"epoch": 4.110872367855608,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00037336779321714376,
|
|
"loss": 5.3738,
|
|
"mean_token_accuracy": 0.18268467783927916,
|
|
"num_tokens": 8699986.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.69266357421875,
|
|
"epoch": 4.115169746454662,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00037306866121544633,
|
|
"loss": 5.3163,
|
|
"mean_token_accuracy": 0.19874982088804244,
|
|
"num_tokens": 8708517.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.76296329498291,
|
|
"epoch": 4.119467125053717,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003727693150709904,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.18448924869298935,
|
|
"num_tokens": 8717556.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.775180006027222,
|
|
"epoch": 4.1237645036527715,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00037246975543742843,
|
|
"loss": 5.3621,
|
|
"mean_token_accuracy": 0.18471318185329438,
|
|
"num_tokens": 8726499.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.7897937297821045,
|
|
"epoch": 4.128061882251826,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000372169982968879,
|
|
"loss": 5.3391,
|
|
"mean_token_accuracy": 0.19769490659236907,
|
|
"num_tokens": 8734449.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.799447870254516,
|
|
"epoch": 4.132359260850881,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0003718699983199252,
|
|
"loss": 5.4169,
|
|
"mean_token_accuracy": 0.18625482618808747,
|
|
"num_tokens": 8744098.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.687497234344482,
|
|
"epoch": 4.136656639449935,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003715698021456137,
|
|
"loss": 5.2403,
|
|
"mean_token_accuracy": 0.20176085084676743,
|
|
"num_tokens": 8753719.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.798142766952514,
|
|
"epoch": 4.1409540180489905,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00037126939510145294,
|
|
"loss": 5.3595,
|
|
"mean_token_accuracy": 0.19173555076122284,
|
|
"num_tokens": 8761883.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.706368494033813,
|
|
"epoch": 4.145251396648045,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003709687778434118,
|
|
"loss": 5.3191,
|
|
"mean_token_accuracy": 0.19367345571517944,
|
|
"num_tokens": 8770594.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.717660665512085,
|
|
"epoch": 4.1495487752471,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003706679510279183,
|
|
"loss": 5.3322,
|
|
"mean_token_accuracy": 0.1913139373064041,
|
|
"num_tokens": 8780314.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.851908397674561,
|
|
"epoch": 4.153846153846154,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003703669153118578,
|
|
"loss": 5.3672,
|
|
"mean_token_accuracy": 0.1877646654844284,
|
|
"num_tokens": 8788832.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.746940755844117,
|
|
"epoch": 4.158143532445209,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00037006567135257216,
|
|
"loss": 5.3745,
|
|
"mean_token_accuracy": 0.18493312001228332,
|
|
"num_tokens": 8797457.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.755455493927002,
|
|
"epoch": 4.162440911044263,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00036976421980785764,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.18980322629213334,
|
|
"num_tokens": 8806760.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.810923004150391,
|
|
"epoch": 4.166738289643318,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003694625613359641,
|
|
"loss": 5.264,
|
|
"mean_token_accuracy": 0.19841254204511644,
|
|
"num_tokens": 8815641.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.728625774383545,
|
|
"epoch": 4.171035668242372,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003691606965955929,
|
|
"loss": 5.3063,
|
|
"mean_token_accuracy": 0.18562910109758377,
|
|
"num_tokens": 8824532.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.817043876647949,
|
|
"epoch": 4.175333046841427,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000368858626245896,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.18077927976846694,
|
|
"num_tokens": 8834896.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.809082460403443,
|
|
"epoch": 4.1796304254404815,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003685563509464744,
|
|
"loss": 5.3836,
|
|
"mean_token_accuracy": 0.19062644392251968,
|
|
"num_tokens": 8844320.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.730808687210083,
|
|
"epoch": 4.183927804039536,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00036825387135737647,
|
|
"loss": 5.2781,
|
|
"mean_token_accuracy": 0.18831727653741837,
|
|
"num_tokens": 8853400.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.736645317077636,
|
|
"epoch": 4.188225182638591,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00036795118813909674,
|
|
"loss": 5.3664,
|
|
"mean_token_accuracy": 0.1930340588092804,
|
|
"num_tokens": 8861930.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.733294582366943,
|
|
"epoch": 4.192522561237645,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00036764830195257437,
|
|
"loss": 5.4453,
|
|
"mean_token_accuracy": 0.18443686068058013,
|
|
"num_tokens": 8871103.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.855679607391357,
|
|
"epoch": 4.1968199398367,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003673452134591918,
|
|
"loss": 5.4556,
|
|
"mean_token_accuracy": 0.1774896889925003,
|
|
"num_tokens": 8879875.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.819746780395508,
|
|
"epoch": 4.201117318435754,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000367041923320773,
|
|
"loss": 5.4289,
|
|
"mean_token_accuracy": 0.17917239516973496,
|
|
"num_tokens": 8888992.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.787553787231445,
|
|
"epoch": 4.205414697034809,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00036673843219958257,
|
|
"loss": 5.4351,
|
|
"mean_token_accuracy": 0.18415016382932664,
|
|
"num_tokens": 8898609.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.777013063430786,
|
|
"epoch": 4.209712075633863,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003664347407583238,
|
|
"loss": 5.4286,
|
|
"mean_token_accuracy": 0.18492118567228316,
|
|
"num_tokens": 8907998.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.754482269287109,
|
|
"epoch": 4.214009454232918,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003661308496601373,
|
|
"loss": 5.3906,
|
|
"mean_token_accuracy": 0.18720373064279555,
|
|
"num_tokens": 8916369.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.810417985916137,
|
|
"epoch": 4.218306832831972,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00036582675956859983,
|
|
"loss": 5.4398,
|
|
"mean_token_accuracy": 0.18979214280843734,
|
|
"num_tokens": 8926482.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.760086631774902,
|
|
"epoch": 4.222604211431027,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.00036552247114772263,
|
|
"loss": 5.2557,
|
|
"mean_token_accuracy": 0.2028429239988327,
|
|
"num_tokens": 8934329.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.750718212127685,
|
|
"epoch": 4.2269015900300815,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036521798506194996,
|
|
"loss": 5.5792,
|
|
"mean_token_accuracy": 0.16662288308143616,
|
|
"num_tokens": 8943606.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.79989538192749,
|
|
"epoch": 4.231198968629136,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00036491330197615775,
|
|
"loss": 5.4319,
|
|
"mean_token_accuracy": 0.18750673532485962,
|
|
"num_tokens": 8953384.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.768419551849365,
|
|
"epoch": 4.235496347228191,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00036460842255565197,
|
|
"loss": 5.3462,
|
|
"mean_token_accuracy": 0.19430523067712785,
|
|
"num_tokens": 8962269.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.792366981506348,
|
|
"epoch": 4.239793725827245,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003643033474661676,
|
|
"loss": 5.3695,
|
|
"mean_token_accuracy": 0.19096554517745973,
|
|
"num_tokens": 8970476.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.765590858459473,
|
|
"epoch": 4.2440911044263,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00036399807737386657,
|
|
"loss": 5.4506,
|
|
"mean_token_accuracy": 0.18137593567371368,
|
|
"num_tokens": 8980322.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.766800260543823,
|
|
"epoch": 4.248388483025354,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003636926129453368,
|
|
"loss": 5.3995,
|
|
"mean_token_accuracy": 0.18808265626430512,
|
|
"num_tokens": 8989575.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.7237457752227785,
|
|
"epoch": 4.252685861624409,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003633869548475904,
|
|
"loss": 5.2917,
|
|
"mean_token_accuracy": 0.19027384221553803,
|
|
"num_tokens": 8998027.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.790141344070435,
|
|
"epoch": 4.256983240223463,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.0003630811037480627,
|
|
"loss": 5.3773,
|
|
"mean_token_accuracy": 0.185378597676754,
|
|
"num_tokens": 9006942.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.791253709793091,
|
|
"epoch": 4.261280618822518,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003627750603146101,
|
|
"loss": 5.3678,
|
|
"mean_token_accuracy": 0.18502196818590164,
|
|
"num_tokens": 9016863.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.759066104888916,
|
|
"epoch": 4.265577997421573,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003624688252155091,
|
|
"loss": 5.4174,
|
|
"mean_token_accuracy": 0.18409577757120132,
|
|
"num_tokens": 9025837.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.820334577560425,
|
|
"epoch": 4.269875376020628,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003621623991194549,
|
|
"loss": 5.3678,
|
|
"mean_token_accuracy": 0.18398485481739044,
|
|
"num_tokens": 9034625.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.809531593322754,
|
|
"epoch": 4.274172754619682,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003618557826955594,
|
|
"loss": 5.4022,
|
|
"mean_token_accuracy": 0.18488193303346634,
|
|
"num_tokens": 9043547.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.807176637649536,
|
|
"epoch": 4.278470133218737,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00036154897661335063,
|
|
"loss": 5.4096,
|
|
"mean_token_accuracy": 0.17959226816892623,
|
|
"num_tokens": 9052857.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.871719217300415,
|
|
"epoch": 4.2827675118177915,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003612419815427702,
|
|
"loss": 5.4076,
|
|
"mean_token_accuracy": 0.18340698778629302,
|
|
"num_tokens": 9062823.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.767588090896607,
|
|
"epoch": 4.287064890416846,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003609347981541726,
|
|
"loss": 5.3928,
|
|
"mean_token_accuracy": 0.1867993012070656,
|
|
"num_tokens": 9071209.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.744509887695313,
|
|
"epoch": 4.291362269015901,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00036062742711832376,
|
|
"loss": 5.4193,
|
|
"mean_token_accuracy": 0.1841732770204544,
|
|
"num_tokens": 9082699.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.786597061157226,
|
|
"epoch": 4.295659647614955,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003603198691063991,
|
|
"loss": 5.3498,
|
|
"mean_token_accuracy": 0.18527137488126755,
|
|
"num_tokens": 9091116.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.295659647614955,
|
|
"eval_entropy": 5.644664139897974,
|
|
"eval_loss": 6.079990386962891,
|
|
"eval_mean_token_accuracy": 0.16622278877945096,
|
|
"eval_num_tokens": 9091116.0,
|
|
"eval_runtime": 2.6418,
|
|
"eval_samples_per_second": 1343.39,
|
|
"eval_steps_per_second": 168.066,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.740189743041992,
|
|
"epoch": 4.29995702621401,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003600121247899824,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.19137779325246812,
|
|
"num_tokens": 9099470.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.692507266998291,
|
|
"epoch": 4.304254404813064,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00035970419484106404,
|
|
"loss": 5.3517,
|
|
"mean_token_accuracy": 0.19267186224460603,
|
|
"num_tokens": 9108367.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.782030725479126,
|
|
"epoch": 4.308551783412119,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003593960799320402,
|
|
"loss": 5.3521,
|
|
"mean_token_accuracy": 0.19390282779932022,
|
|
"num_tokens": 9116560.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.794342994689941,
|
|
"epoch": 4.312849162011173,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003590877807357107,
|
|
"loss": 5.4196,
|
|
"mean_token_accuracy": 0.18967381864786148,
|
|
"num_tokens": 9124634.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.809728908538818,
|
|
"epoch": 4.317146540610228,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003587792979252776,
|
|
"loss": 5.4103,
|
|
"mean_token_accuracy": 0.18608468323945998,
|
|
"num_tokens": 9133246.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.769702386856079,
|
|
"epoch": 4.321443919209282,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003584706321743442,
|
|
"loss": 5.3763,
|
|
"mean_token_accuracy": 0.1830572485923767,
|
|
"num_tokens": 9142374.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.809864568710327,
|
|
"epoch": 4.325741297808337,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000358161784156913,
|
|
"loss": 5.4148,
|
|
"mean_token_accuracy": 0.1828118771314621,
|
|
"num_tokens": 9151657.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.723864984512329,
|
|
"epoch": 4.3300386764073915,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00035785275454738456,
|
|
"loss": 5.4721,
|
|
"mean_token_accuracy": 0.18104234486818313,
|
|
"num_tokens": 9159897.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.696597766876221,
|
|
"epoch": 4.334336055006446,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00035754354402055635,
|
|
"loss": 5.4615,
|
|
"mean_token_accuracy": 0.18050874769687653,
|
|
"num_tokens": 9169798.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.897407531738281,
|
|
"epoch": 4.338633433605501,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003572341532516202,
|
|
"loss": 5.5435,
|
|
"mean_token_accuracy": 0.16937263458967208,
|
|
"num_tokens": 9179107.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.817531108856201,
|
|
"epoch": 4.342930812204555,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003569245829161622,
|
|
"loss": 5.4084,
|
|
"mean_token_accuracy": 0.18091799616813659,
|
|
"num_tokens": 9188295.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.752144861221313,
|
|
"epoch": 4.34722819080361,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035661483369016004,
|
|
"loss": 5.3673,
|
|
"mean_token_accuracy": 0.19427870064973832,
|
|
"num_tokens": 9197603.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.729626655578613,
|
|
"epoch": 4.351525569402664,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003563049062499822,
|
|
"loss": 5.3927,
|
|
"mean_token_accuracy": 0.18378866463899612,
|
|
"num_tokens": 9206413.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.824339532852173,
|
|
"epoch": 4.355822948001719,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0003559948012723865,
|
|
"loss": 5.4666,
|
|
"mean_token_accuracy": 0.1768575355410576,
|
|
"num_tokens": 9215376.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.855458402633667,
|
|
"epoch": 4.360120326600773,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003556845194345181,
|
|
"loss": 5.4706,
|
|
"mean_token_accuracy": 0.1757975623011589,
|
|
"num_tokens": 9224418.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.752833032608033,
|
|
"epoch": 4.364417705199828,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003553740614139086,
|
|
"loss": 5.3542,
|
|
"mean_token_accuracy": 0.18811701983213425,
|
|
"num_tokens": 9235083.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.7536077976226805,
|
|
"epoch": 4.368715083798882,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003550634278884742,
|
|
"loss": 5.3288,
|
|
"mean_token_accuracy": 0.2002038314938545,
|
|
"num_tokens": 9244576.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.741670703887939,
|
|
"epoch": 4.373012462397937,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035475261953651433,
|
|
"loss": 5.3835,
|
|
"mean_token_accuracy": 0.190900257229805,
|
|
"num_tokens": 9253969.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.794024991989136,
|
|
"epoch": 4.3773098409969915,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035444163703671026,
|
|
"loss": 5.4962,
|
|
"mean_token_accuracy": 0.18178885281085969,
|
|
"num_tokens": 9264723.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.805999040603638,
|
|
"epoch": 4.381607219596046,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00035413048106812357,
|
|
"loss": 5.343,
|
|
"mean_token_accuracy": 0.18907343596220016,
|
|
"num_tokens": 9273863.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.762822484970092,
|
|
"epoch": 4.385904598195101,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00035381915231019425,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.1856116086244583,
|
|
"num_tokens": 9282633.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.801623344421387,
|
|
"epoch": 4.390201976794156,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003535076514427401,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.17852161824703217,
|
|
"num_tokens": 9291575.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.801022386550903,
|
|
"epoch": 4.39449935539321,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00035319597914595436,
|
|
"loss": 5.4007,
|
|
"mean_token_accuracy": 0.18363688737154008,
|
|
"num_tokens": 9299883.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.773863220214844,
|
|
"epoch": 4.398796733992265,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003528841361004049,
|
|
"loss": 5.395,
|
|
"mean_token_accuracy": 0.1873690128326416,
|
|
"num_tokens": 9308828.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.767456007003784,
|
|
"epoch": 4.40309411259132,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003525721229870323,
|
|
"loss": 5.4414,
|
|
"mean_token_accuracy": 0.1802488774061203,
|
|
"num_tokens": 9318458.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.763554477691651,
|
|
"epoch": 4.407391491190374,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00035225994048714823,
|
|
"loss": 5.3697,
|
|
"mean_token_accuracy": 0.1885476216673851,
|
|
"num_tokens": 9327811.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.710635662078857,
|
|
"epoch": 4.411688869789429,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003519475892824348,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.19102655947208405,
|
|
"num_tokens": 9336169.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.7136293888092045,
|
|
"epoch": 4.415986248388483,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003516350700549419,
|
|
"loss": 5.3854,
|
|
"mean_token_accuracy": 0.19286593645811081,
|
|
"num_tokens": 9345141.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.758143186569214,
|
|
"epoch": 4.420283626987538,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00035132238348708697,
|
|
"loss": 5.2838,
|
|
"mean_token_accuracy": 0.19995164573192598,
|
|
"num_tokens": 9353982.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.772878789901734,
|
|
"epoch": 4.424581005586592,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00035100953026165224,
|
|
"loss": 5.3701,
|
|
"mean_token_accuracy": 0.19822736084461212,
|
|
"num_tokens": 9362991.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.658030462265015,
|
|
"epoch": 4.428878384185647,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003506965110617841,
|
|
"loss": 5.3794,
|
|
"mean_token_accuracy": 0.18665509223937987,
|
|
"num_tokens": 9372310.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.765686464309693,
|
|
"epoch": 4.4331757627847015,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0003503833265709915,
|
|
"loss": 5.4097,
|
|
"mean_token_accuracy": 0.1783752828836441,
|
|
"num_tokens": 9381268.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.853574323654175,
|
|
"epoch": 4.437473141383756,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00035006997747314404,
|
|
"loss": 5.3805,
|
|
"mean_token_accuracy": 0.19683203101158142,
|
|
"num_tokens": 9389670.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.697288703918457,
|
|
"epoch": 4.441770519982811,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00034975646445247106,
|
|
"loss": 5.3389,
|
|
"mean_token_accuracy": 0.1868162587285042,
|
|
"num_tokens": 9397520.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.769460296630859,
|
|
"epoch": 4.446067898581865,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003494427881935596,
|
|
"loss": 5.3654,
|
|
"mean_token_accuracy": 0.18655847311019896,
|
|
"num_tokens": 9406848.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.763192558288575,
|
|
"epoch": 4.45036527718092,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00034912894938135325,
|
|
"loss": 5.3685,
|
|
"mean_token_accuracy": 0.18144838362932206,
|
|
"num_tokens": 9415820.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.6896614074707035,
|
|
"epoch": 4.454662655779974,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003488149487011506,
|
|
"loss": 5.3303,
|
|
"mean_token_accuracy": 0.19242522716522217,
|
|
"num_tokens": 9424907.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.742438936233521,
|
|
"epoch": 4.458960034379029,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00034850078683860346,
|
|
"loss": 5.4064,
|
|
"mean_token_accuracy": 0.19099208414554597,
|
|
"num_tokens": 9433882.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.847838592529297,
|
|
"epoch": 4.463257412978083,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.0003481864644797159,
|
|
"loss": 5.39,
|
|
"mean_token_accuracy": 0.1870894268155098,
|
|
"num_tokens": 9443544.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.779017782211303,
|
|
"epoch": 4.467554791577138,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003478719823108424,
|
|
"loss": 5.4259,
|
|
"mean_token_accuracy": 0.17777402102947235,
|
|
"num_tokens": 9452473.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.728008365631103,
|
|
"epoch": 4.471852170176192,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00034755734101868613,
|
|
"loss": 5.4058,
|
|
"mean_token_accuracy": 0.18134617805480957,
|
|
"num_tokens": 9461392.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.717585563659668,
|
|
"epoch": 4.476149548775247,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034724254129029795,
|
|
"loss": 5.3832,
|
|
"mean_token_accuracy": 0.19526348114013672,
|
|
"num_tokens": 9470087.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.722295761108398,
|
|
"epoch": 4.4804469273743015,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003469275838130748,
|
|
"loss": 5.3615,
|
|
"mean_token_accuracy": 0.19543582797050477,
|
|
"num_tokens": 9479654.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.803108739852905,
|
|
"epoch": 4.484744305973356,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003466124692747577,
|
|
"loss": 5.4219,
|
|
"mean_token_accuracy": 0.1773220792412758,
|
|
"num_tokens": 9488116.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.7404860496521,
|
|
"epoch": 4.489041684572411,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00034629719836343106,
|
|
"loss": 5.347,
|
|
"mean_token_accuracy": 0.19328135401010513,
|
|
"num_tokens": 9496853.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.749950408935547,
|
|
"epoch": 4.493339063171465,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003459817717675203,
|
|
"loss": 5.3532,
|
|
"mean_token_accuracy": 0.1924394577741623,
|
|
"num_tokens": 9507085.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.745312309265136,
|
|
"epoch": 4.49763644177052,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003456661901757913,
|
|
"loss": 5.4143,
|
|
"mean_token_accuracy": 0.18019952327013017,
|
|
"num_tokens": 9516393.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.770188713073731,
|
|
"epoch": 4.501933820369574,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00034535045427734796,
|
|
"loss": 5.4133,
|
|
"mean_token_accuracy": 0.18343990445137023,
|
|
"num_tokens": 9525306.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.778894519805908,
|
|
"epoch": 4.506231198968629,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003450345647616313,
|
|
"loss": 5.4242,
|
|
"mean_token_accuracy": 0.1856287822127342,
|
|
"num_tokens": 9534401.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.78074164390564,
|
|
"epoch": 4.510528577567683,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003447185223184177,
|
|
"loss": 5.4687,
|
|
"mean_token_accuracy": 0.18452950119972228,
|
|
"num_tokens": 9544651.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.740122938156128,
|
|
"epoch": 4.514825956166739,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00034440232763781765,
|
|
"loss": 5.3652,
|
|
"mean_token_accuracy": 0.19316569119691848,
|
|
"num_tokens": 9552903.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.712041854858398,
|
|
"epoch": 4.519123334765792,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000344085981410274,
|
|
"loss": 5.3789,
|
|
"mean_token_accuracy": 0.1908051148056984,
|
|
"num_tokens": 9561129.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.779702186584473,
|
|
"epoch": 4.523420713364848,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00034376948432656036,
|
|
"loss": 5.3819,
|
|
"mean_token_accuracy": 0.1827898293733597,
|
|
"num_tokens": 9570075.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.777800035476685,
|
|
"epoch": 4.527718091963902,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003434528370777798,
|
|
"loss": 5.3345,
|
|
"mean_token_accuracy": 0.19055496007204056,
|
|
"num_tokens": 9578352.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.721065044403076,
|
|
"epoch": 4.532015470562957,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00034313604035536344,
|
|
"loss": 5.4205,
|
|
"mean_token_accuracy": 0.19161821156740189,
|
|
"num_tokens": 9587049.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.801066732406616,
|
|
"epoch": 4.5363128491620115,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003428190948510687,
|
|
"loss": 5.4099,
|
|
"mean_token_accuracy": 0.18362002968788146,
|
|
"num_tokens": 9595856.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.7225730419158936,
|
|
"epoch": 4.540610227761066,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003425020012569778,
|
|
"loss": 5.3382,
|
|
"mean_token_accuracy": 0.19539454579353333,
|
|
"num_tokens": 9604258.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.788104104995727,
|
|
"epoch": 4.544907606360121,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00034218476026549665,
|
|
"loss": 5.4113,
|
|
"mean_token_accuracy": 0.18487139642238617,
|
|
"num_tokens": 9613961.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.85221643447876,
|
|
"epoch": 4.549204984959175,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003418673725693524,
|
|
"loss": 5.4787,
|
|
"mean_token_accuracy": 0.1759480580687523,
|
|
"num_tokens": 9623440.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.803034877777099,
|
|
"epoch": 4.55350236355823,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003415498388615932,
|
|
"loss": 5.4276,
|
|
"mean_token_accuracy": 0.18098911494016648,
|
|
"num_tokens": 9632988.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.73315463066101,
|
|
"epoch": 4.557799742157284,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003412321598355857,
|
|
"loss": 5.4347,
|
|
"mean_token_accuracy": 0.19001236706972122,
|
|
"num_tokens": 9641479.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.798529863357544,
|
|
"epoch": 4.562097120756339,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003409143361850139,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.1868632197380066,
|
|
"num_tokens": 9651088.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.755837440490723,
|
|
"epoch": 4.566394499355393,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003405963686038775,
|
|
"loss": 5.3245,
|
|
"mean_token_accuracy": 0.18831714242696762,
|
|
"num_tokens": 9659969.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.78913140296936,
|
|
"epoch": 4.570691877954448,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003402782577864908,
|
|
"loss": 5.4639,
|
|
"mean_token_accuracy": 0.18091929256916045,
|
|
"num_tokens": 9669509.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.68860125541687,
|
|
"epoch": 4.574989256553502,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00033996000442748056,
|
|
"loss": 5.3515,
|
|
"mean_token_accuracy": 0.1889285609126091,
|
|
"num_tokens": 9677758.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.7625631332397464,
|
|
"epoch": 4.579286635152557,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033964160922178495,
|
|
"loss": 5.4611,
|
|
"mean_token_accuracy": 0.18348290920257568,
|
|
"num_tokens": 9686622.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.747465467453003,
|
|
"epoch": 4.5835840137516115,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003393230728646518,
|
|
"loss": 5.363,
|
|
"mean_token_accuracy": 0.18685472011566162,
|
|
"num_tokens": 9694726.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.702461862564087,
|
|
"epoch": 4.587881392350666,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00033900439605163724,
|
|
"loss": 5.3631,
|
|
"mean_token_accuracy": 0.18905410319566726,
|
|
"num_tokens": 9703137.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.7536420822143555,
|
|
"epoch": 4.592178770949721,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00033868557947860407,
|
|
"loss": 5.382,
|
|
"mean_token_accuracy": 0.18702416568994523,
|
|
"num_tokens": 9711818.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.705536031723023,
|
|
"epoch": 4.596476149548775,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00033836662384172014,
|
|
"loss": 5.3865,
|
|
"mean_token_accuracy": 0.19549935311079025,
|
|
"num_tokens": 9720678.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.779536294937134,
|
|
"epoch": 4.60077352814783,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003380475298374573,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.1852946788072586,
|
|
"num_tokens": 9729704.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.731397867202759,
|
|
"epoch": 4.605070906746884,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000337728298162589,
|
|
"loss": 5.4147,
|
|
"mean_token_accuracy": 0.19023727029561996,
|
|
"num_tokens": 9739223.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.76360387802124,
|
|
"epoch": 4.609368285345939,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033740892951418993,
|
|
"loss": 5.4242,
|
|
"mean_token_accuracy": 0.194333016872406,
|
|
"num_tokens": 9748905.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.753440904617309,
|
|
"epoch": 4.613665663944993,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003370894245896333,
|
|
"loss": 5.4664,
|
|
"mean_token_accuracy": 0.18534409552812575,
|
|
"num_tokens": 9758156.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.717299509048462,
|
|
"epoch": 4.617963042544048,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00033676978408659047,
|
|
"loss": 5.4276,
|
|
"mean_token_accuracy": 0.18554221242666244,
|
|
"num_tokens": 9767547.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.7771378517150875,
|
|
"epoch": 4.622260421143102,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003364500087030283,
|
|
"loss": 5.5263,
|
|
"mean_token_accuracy": 0.1819859430193901,
|
|
"num_tokens": 9777448.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.864838886260986,
|
|
"epoch": 4.626557799742157,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00033613009913720845,
|
|
"loss": 5.4652,
|
|
"mean_token_accuracy": 0.18829712122678757,
|
|
"num_tokens": 9785856.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.8268232345581055,
|
|
"epoch": 4.6308551783412115,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00033581005608768563,
|
|
"loss": 5.4533,
|
|
"mean_token_accuracy": 0.1869758665561676,
|
|
"num_tokens": 9794806.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.832350397109986,
|
|
"epoch": 4.635152556940266,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003354898802533058,
|
|
"loss": 5.3924,
|
|
"mean_token_accuracy": 0.1853599190711975,
|
|
"num_tokens": 9804105.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.8090879917144775,
|
|
"epoch": 4.6394499355393215,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003351695723332051,
|
|
"loss": 5.5249,
|
|
"mean_token_accuracy": 0.17509887218475342,
|
|
"num_tokens": 9812875.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.713995027542114,
|
|
"epoch": 4.643747314138375,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00033484913302680807,
|
|
"loss": 5.2843,
|
|
"mean_token_accuracy": 0.19741771072149278,
|
|
"num_tokens": 9821542.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.716259288787842,
|
|
"epoch": 4.648044692737431,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033452856303382595,
|
|
"loss": 5.309,
|
|
"mean_token_accuracy": 0.2005496770143509,
|
|
"num_tokens": 9830924.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.734077405929566,
|
|
"epoch": 4.652342071336484,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003342078630542555,
|
|
"loss": 5.3993,
|
|
"mean_token_accuracy": 0.19129912108182906,
|
|
"num_tokens": 9839967.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.769095849990845,
|
|
"epoch": 4.65663944993554,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033388703378837737,
|
|
"loss": 5.439,
|
|
"mean_token_accuracy": 0.17480305582284927,
|
|
"num_tokens": 9849129.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.781463861465454,
|
|
"epoch": 4.660936828534594,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003335660759367544,
|
|
"loss": 5.4282,
|
|
"mean_token_accuracy": 0.18665772974491118,
|
|
"num_tokens": 9858303.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.835739946365356,
|
|
"epoch": 4.665234207133649,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033324499020023025,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.17668160498142244,
|
|
"num_tokens": 9866886.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.734031867980957,
|
|
"epoch": 4.669531585732703,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003329237772799277,
|
|
"loss": 5.3355,
|
|
"mean_token_accuracy": 0.1989918455481529,
|
|
"num_tokens": 9875474.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.67003698348999,
|
|
"epoch": 4.673828964331758,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003326024378772477,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.1989797681570053,
|
|
"num_tokens": 9885104.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.776555681228638,
|
|
"epoch": 4.678126342930812,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003322809726938667,
|
|
"loss": 5.4567,
|
|
"mean_token_accuracy": 0.18246063590049744,
|
|
"num_tokens": 9894698.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.78788537979126,
|
|
"epoch": 4.682423721529867,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00033195938243173645,
|
|
"loss": 5.3798,
|
|
"mean_token_accuracy": 0.1911837249994278,
|
|
"num_tokens": 9903850.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.801269245147705,
|
|
"epoch": 4.6867211001289215,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.0003316376677930814,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.17792850285768508,
|
|
"num_tokens": 9912823.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.75900673866272,
|
|
"epoch": 4.691018478727976,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003313158294803977,
|
|
"loss": 5.4112,
|
|
"mean_token_accuracy": 0.18479508608579637,
|
|
"num_tokens": 9922279.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.784261798858642,
|
|
"epoch": 4.695315857327031,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00033099386819645176,
|
|
"loss": 5.4949,
|
|
"mean_token_accuracy": 0.18499183505773545,
|
|
"num_tokens": 9931646.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.812231826782226,
|
|
"epoch": 4.699613235926085,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003306717846442782,
|
|
"loss": 5.462,
|
|
"mean_token_accuracy": 0.18356119096279144,
|
|
"num_tokens": 9940672.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.698050594329834,
|
|
"epoch": 4.70391061452514,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003303495795271788,
|
|
"loss": 5.3094,
|
|
"mean_token_accuracy": 0.1928895264863968,
|
|
"num_tokens": 9948954.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.813793039321899,
|
|
"epoch": 4.708207993124194,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033002725354872075,
|
|
"loss": 5.4857,
|
|
"mean_token_accuracy": 0.18250587880611419,
|
|
"num_tokens": 9959132.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.811142063140869,
|
|
"epoch": 4.712505371723249,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032970480741273514,
|
|
"loss": 5.502,
|
|
"mean_token_accuracy": 0.18048388361930848,
|
|
"num_tokens": 9969143.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.770669174194336,
|
|
"epoch": 4.716802750322303,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003293822418233155,
|
|
"loss": 5.4437,
|
|
"mean_token_accuracy": 0.18108683377504348,
|
|
"num_tokens": 9978462.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.838630199432373,
|
|
"epoch": 4.721100128921358,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003290595574848161,
|
|
"loss": 5.4805,
|
|
"mean_token_accuracy": 0.17529057562351227,
|
|
"num_tokens": 9987543.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.84181227684021,
|
|
"epoch": 4.725397507520412,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003287367551018505,
|
|
"loss": 5.4928,
|
|
"mean_token_accuracy": 0.17725082337856293,
|
|
"num_tokens": 9996573.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.725397507520412,
|
|
"eval_entropy": 5.631523649434786,
|
|
"eval_loss": 6.047227382659912,
|
|
"eval_mean_token_accuracy": 0.16825610127400709,
|
|
"eval_num_tokens": 9996573.0,
|
|
"eval_runtime": 2.6495,
|
|
"eval_samples_per_second": 1339.485,
|
|
"eval_steps_per_second": 167.577,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.789327239990234,
|
|
"epoch": 4.729694886119467,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.0003284138353792903,
|
|
"loss": 5.3912,
|
|
"mean_token_accuracy": 0.18088296204805374,
|
|
"num_tokens": 10005954.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.678737306594849,
|
|
"epoch": 4.7339922647185215,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003280907990222628,
|
|
"loss": 5.3537,
|
|
"mean_token_accuracy": 0.1802771881222725,
|
|
"num_tokens": 10014596.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.670832014083862,
|
|
"epoch": 4.738289643317576,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00032776764673615055,
|
|
"loss": 5.4102,
|
|
"mean_token_accuracy": 0.18394816815853118,
|
|
"num_tokens": 10023787.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.755765628814697,
|
|
"epoch": 4.742587021916631,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003274443792265888,
|
|
"loss": 5.3985,
|
|
"mean_token_accuracy": 0.18589506596326827,
|
|
"num_tokens": 10031470.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.7581878185272215,
|
|
"epoch": 4.746884400515685,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00032712099719946474,
|
|
"loss": 5.3326,
|
|
"mean_token_accuracy": 0.19611956179141998,
|
|
"num_tokens": 10039558.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.740596580505371,
|
|
"epoch": 4.75118177911474,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00032679750136091533,
|
|
"loss": 5.4038,
|
|
"mean_token_accuracy": 0.18175314366817474,
|
|
"num_tokens": 10049244.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.7172479152679445,
|
|
"epoch": 4.755479157713794,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003264738924173262,
|
|
"loss": 5.4035,
|
|
"mean_token_accuracy": 0.17695202827453613,
|
|
"num_tokens": 10058109.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.755099821090698,
|
|
"epoch": 4.759776536312849,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00032615017107533,
|
|
"loss": 5.3543,
|
|
"mean_token_accuracy": 0.19952582716941833,
|
|
"num_tokens": 10066824.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.761263084411621,
|
|
"epoch": 4.764073914911903,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003258263380418047,
|
|
"loss": 5.3509,
|
|
"mean_token_accuracy": 0.19581466168165207,
|
|
"num_tokens": 10075399.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.805514812469482,
|
|
"epoch": 4.768371293510958,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00032550239402387226,
|
|
"loss": 5.3855,
|
|
"mean_token_accuracy": 0.19303188025951384,
|
|
"num_tokens": 10083846.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.774110555648804,
|
|
"epoch": 4.772668672110013,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00032517833972889695,
|
|
"loss": 5.3829,
|
|
"mean_token_accuracy": 0.18658793717622757,
|
|
"num_tokens": 10092238.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.733133602142334,
|
|
"epoch": 4.776966050709067,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00032485417586448375,
|
|
"loss": 5.3621,
|
|
"mean_token_accuracy": 0.1931225672364235,
|
|
"num_tokens": 10101195.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.731488990783691,
|
|
"epoch": 4.781263429308122,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000324529903138477,
|
|
"loss": 5.4526,
|
|
"mean_token_accuracy": 0.18415073156356812,
|
|
"num_tokens": 10110226.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.641093969345093,
|
|
"epoch": 4.785560807907177,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003242055222589587,
|
|
"loss": 5.3502,
|
|
"mean_token_accuracy": 0.19548950344324112,
|
|
"num_tokens": 10119263.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.837341403961181,
|
|
"epoch": 4.7898581865062315,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000323881033934247,
|
|
"loss": 5.3552,
|
|
"mean_token_accuracy": 0.18924368172883987,
|
|
"num_tokens": 10128548.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.77348518371582,
|
|
"epoch": 4.794155565105286,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00032355643887289486,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.18653484135866166,
|
|
"num_tokens": 10137455.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.682504034042358,
|
|
"epoch": 4.798452943704341,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003232317377836881,
|
|
"loss": 5.3547,
|
|
"mean_token_accuracy": 0.1950716495513916,
|
|
"num_tokens": 10146469.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.712986850738526,
|
|
"epoch": 4.802750322303395,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000322906931375644,
|
|
"loss": 5.3907,
|
|
"mean_token_accuracy": 0.18491413444280624,
|
|
"num_tokens": 10155898.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.784373331069946,
|
|
"epoch": 4.80704770090245,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032258202035801,
|
|
"loss": 5.4543,
|
|
"mean_token_accuracy": 0.18605074882507325,
|
|
"num_tokens": 10165099.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.776973581314087,
|
|
"epoch": 4.811345079501504,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000322257005440262,
|
|
"loss": 5.3823,
|
|
"mean_token_accuracy": 0.1842188060283661,
|
|
"num_tokens": 10174240.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.737133121490478,
|
|
"epoch": 4.815642458100559,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003219318873321025,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.1906519427895546,
|
|
"num_tokens": 10183389.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.802797031402588,
|
|
"epoch": 4.819939836699613,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00032160666674345954,
|
|
"loss": 5.3998,
|
|
"mean_token_accuracy": 0.1901124358177185,
|
|
"num_tokens": 10191749.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.6718216896057125,
|
|
"epoch": 4.824237215298668,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032128134438448504,
|
|
"loss": 5.403,
|
|
"mean_token_accuracy": 0.18799017518758773,
|
|
"num_tokens": 10200759.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.745155191421508,
|
|
"epoch": 4.828534593897722,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032095592096555284,
|
|
"loss": 5.4024,
|
|
"mean_token_accuracy": 0.18890567421913146,
|
|
"num_tokens": 10211141.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.824276351928711,
|
|
"epoch": 4.832831972496777,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003206303971972577,
|
|
"loss": 5.4038,
|
|
"mean_token_accuracy": 0.18879446834325792,
|
|
"num_tokens": 10221412.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.754834175109863,
|
|
"epoch": 4.8371293510958315,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003203047737904134,
|
|
"loss": 5.3973,
|
|
"mean_token_accuracy": 0.18383289575576783,
|
|
"num_tokens": 10229894.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.745780658721924,
|
|
"epoch": 4.841426729694886,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031997905145605135,
|
|
"loss": 5.5037,
|
|
"mean_token_accuracy": 0.17905805706977845,
|
|
"num_tokens": 10239636.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.7630202293396,
|
|
"epoch": 4.845724108293941,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00031965323090541874,
|
|
"loss": 5.4406,
|
|
"mean_token_accuracy": 0.18748103380203246,
|
|
"num_tokens": 10248694.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.783816862106323,
|
|
"epoch": 4.850021486892995,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0003193273128499777,
|
|
"loss": 5.3962,
|
|
"mean_token_accuracy": 0.18428648114204407,
|
|
"num_tokens": 10258102.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.685248327255249,
|
|
"epoch": 4.85431886549205,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00031900129800140287,
|
|
"loss": 5.3316,
|
|
"mean_token_accuracy": 0.19054149687290192,
|
|
"num_tokens": 10266752.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.692747020721436,
|
|
"epoch": 4.858616244091104,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00031867518707158027,
|
|
"loss": 5.3512,
|
|
"mean_token_accuracy": 0.18527395278215408,
|
|
"num_tokens": 10276199.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.746193885803223,
|
|
"epoch": 4.862913622690159,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000318348980772606,
|
|
"loss": 5.3626,
|
|
"mean_token_accuracy": 0.19036828279495238,
|
|
"num_tokens": 10285160.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.759525966644287,
|
|
"epoch": 4.867211001289213,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031802267981678414,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.1840071678161621,
|
|
"num_tokens": 10295266.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.707964468002319,
|
|
"epoch": 4.871508379888268,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00031769628491662563,
|
|
"loss": 5.4784,
|
|
"mean_token_accuracy": 0.1852257788181305,
|
|
"num_tokens": 10305155.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.819545745849609,
|
|
"epoch": 4.8758057584873224,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00031736979678484634,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.1751335844397545,
|
|
"num_tokens": 10315052.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.880706644058227,
|
|
"epoch": 4.880103137086377,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00031704321613436597,
|
|
"loss": 5.4187,
|
|
"mean_token_accuracy": 0.1832166850566864,
|
|
"num_tokens": 10324032.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.728805828094482,
|
|
"epoch": 4.8844005156854315,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003167165436783061,
|
|
"loss": 5.4285,
|
|
"mean_token_accuracy": 0.19092569053173064,
|
|
"num_tokens": 10333274.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.641194152832031,
|
|
"epoch": 4.888697894284486,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00031638978012998875,
|
|
"loss": 5.3876,
|
|
"mean_token_accuracy": 0.18218838423490524,
|
|
"num_tokens": 10341873.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.829899883270263,
|
|
"epoch": 4.892995272883541,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000316062926202935,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.18971192687749863,
|
|
"num_tokens": 10350579.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.7446722984313965,
|
|
"epoch": 4.897292651482596,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003157359826108632,
|
|
"loss": 5.4302,
|
|
"mean_token_accuracy": 0.1906385079026222,
|
|
"num_tokens": 10359615.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.665975332260132,
|
|
"epoch": 4.90159003008165,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00031540895006768727,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.18720715045928954,
|
|
"num_tokens": 10367856.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.757908296585083,
|
|
"epoch": 4.905887408680705,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003150818292875158,
|
|
"loss": 5.4418,
|
|
"mean_token_accuracy": 0.18959396481513976,
|
|
"num_tokens": 10377471.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.798106002807617,
|
|
"epoch": 4.91018478727976,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003147546209846497,
|
|
"loss": 5.4842,
|
|
"mean_token_accuracy": 0.18160897940397264,
|
|
"num_tokens": 10387162.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.7933940410614015,
|
|
"epoch": 4.914482165878814,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003144273258735812,
|
|
"loss": 5.3742,
|
|
"mean_token_accuracy": 0.19006772041320802,
|
|
"num_tokens": 10395995.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.755883026123047,
|
|
"epoch": 4.918779544477869,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003140999446689919,
|
|
"loss": 5.442,
|
|
"mean_token_accuracy": 0.18640946298837663,
|
|
"num_tokens": 10405151.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.811358070373535,
|
|
"epoch": 4.923076923076923,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003137724780857516,
|
|
"loss": 5.3782,
|
|
"mean_token_accuracy": 0.18572515547275542,
|
|
"num_tokens": 10414811.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.785015487670899,
|
|
"epoch": 4.927374301675978,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00031344492683891634,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.18111189752817153,
|
|
"num_tokens": 10423822.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.633029937744141,
|
|
"epoch": 4.931671680275032,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003131172916437272,
|
|
"loss": 5.2716,
|
|
"mean_token_accuracy": 0.20237637013196946,
|
|
"num_tokens": 10433014.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.734432935714722,
|
|
"epoch": 4.935969058874087,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031278957321560845,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.19022650122642518,
|
|
"num_tokens": 10442307.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.7562150955200195,
|
|
"epoch": 4.9402664374731415,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00031246177227016615,
|
|
"loss": 5.3338,
|
|
"mean_token_accuracy": 0.1968904510140419,
|
|
"num_tokens": 10451780.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.792382621765137,
|
|
"epoch": 4.944563816072196,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00031213388952318653,
|
|
"loss": 5.4116,
|
|
"mean_token_accuracy": 0.18679167330265045,
|
|
"num_tokens": 10461441.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.759060335159302,
|
|
"epoch": 4.948861194671251,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003118059256906345,
|
|
"loss": 5.4012,
|
|
"mean_token_accuracy": 0.18667513579130174,
|
|
"num_tokens": 10470872.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.668685722351074,
|
|
"epoch": 4.953158573270305,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00031147788148865204,
|
|
"loss": 5.3512,
|
|
"mean_token_accuracy": 0.19128614217042922,
|
|
"num_tokens": 10480414.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.761059188842774,
|
|
"epoch": 4.95745595186936,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003111497576335564,
|
|
"loss": 5.5057,
|
|
"mean_token_accuracy": 0.18111792355775833,
|
|
"num_tokens": 10489837.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.764217281341553,
|
|
"epoch": 4.961753330468414,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003108215548418391,
|
|
"loss": 5.439,
|
|
"mean_token_accuracy": 0.19322182536125182,
|
|
"num_tokens": 10498757.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.749614334106445,
|
|
"epoch": 4.966050709067469,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003104932738301637,
|
|
"loss": 5.3622,
|
|
"mean_token_accuracy": 0.19543524831533432,
|
|
"num_tokens": 10509057.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.792864561080933,
|
|
"epoch": 4.970348087666523,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00031016491531536477,
|
|
"loss": 5.4843,
|
|
"mean_token_accuracy": 0.18161966353654863,
|
|
"num_tokens": 10518207.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.691180658340454,
|
|
"epoch": 4.974645466265578,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003098364800144462,
|
|
"loss": 5.4412,
|
|
"mean_token_accuracy": 0.19089352041482927,
|
|
"num_tokens": 10527564.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.748730802536011,
|
|
"epoch": 4.9789428448646325,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003095079686445792,
|
|
"loss": 5.3725,
|
|
"mean_token_accuracy": 0.18982030749320983,
|
|
"num_tokens": 10536317.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.839184522628784,
|
|
"epoch": 4.983240223463687,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00030917938192310146,
|
|
"loss": 5.4614,
|
|
"mean_token_accuracy": 0.17821027040481568,
|
|
"num_tokens": 10545451.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.775116205215454,
|
|
"epoch": 4.9875376020627415,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00030885072056751494,
|
|
"loss": 5.4518,
|
|
"mean_token_accuracy": 0.1820235088467598,
|
|
"num_tokens": 10554244.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.737451791763306,
|
|
"epoch": 4.991834980661796,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00030852198529548476,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.19195319563150406,
|
|
"num_tokens": 10562815.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.718679523468017,
|
|
"epoch": 4.996132359260851,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003081931768248373,
|
|
"loss": 5.3513,
|
|
"mean_token_accuracy": 0.18969926089048386,
|
|
"num_tokens": 10571136.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.750847498575847,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.0003078642958735588,
|
|
"loss": 5.3201,
|
|
"mean_token_accuracy": 0.18818063371711308,
|
|
"num_tokens": 10579660.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.672137594223022,
|
|
"epoch": 5.0042973785990545,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00030753534315979393,
|
|
"loss": 5.1179,
|
|
"mean_token_accuracy": 0.20628433227539061,
|
|
"num_tokens": 10587594.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.652815532684326,
|
|
"epoch": 5.008594757198109,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003072063194018438,
|
|
"loss": 5.2439,
|
|
"mean_token_accuracy": 0.1979988232254982,
|
|
"num_tokens": 10596694.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.7318213939666744,
|
|
"epoch": 5.012892135797164,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003068772253181648,
|
|
"loss": 5.1648,
|
|
"mean_token_accuracy": 0.20323600023984909,
|
|
"num_tokens": 10605367.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.704654121398926,
|
|
"epoch": 5.017189514396218,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0003065480616273671,
|
|
"loss": 5.1535,
|
|
"mean_token_accuracy": 0.1988356202840805,
|
|
"num_tokens": 10614094.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.755032110214233,
|
|
"epoch": 5.021486892995273,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003062188290482123,
|
|
"loss": 5.221,
|
|
"mean_token_accuracy": 0.2003185749053955,
|
|
"num_tokens": 10624336.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.782719278335572,
|
|
"epoch": 5.025784271594327,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00030588952829961304,
|
|
"loss": 5.2448,
|
|
"mean_token_accuracy": 0.19757871627807616,
|
|
"num_tokens": 10633461.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.732550716400146,
|
|
"epoch": 5.030081650193382,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003055601601006303,
|
|
"loss": 5.2151,
|
|
"mean_token_accuracy": 0.1955297753214836,
|
|
"num_tokens": 10641604.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.6935615062713625,
|
|
"epoch": 5.034379028792436,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003052307251704728,
|
|
"loss": 5.2211,
|
|
"mean_token_accuracy": 0.19739024043083192,
|
|
"num_tokens": 10650628.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.6747438430786135,
|
|
"epoch": 5.038676407391491,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003049012242284946,
|
|
"loss": 5.1139,
|
|
"mean_token_accuracy": 0.2092691496014595,
|
|
"num_tokens": 10659887.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.7882849216461185,
|
|
"epoch": 5.0429737859905455,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003045716579941941,
|
|
"loss": 5.318,
|
|
"mean_token_accuracy": 0.18973567336797714,
|
|
"num_tokens": 10669401.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.7535899639129635,
|
|
"epoch": 5.0472711645896,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00030424202718721215,
|
|
"loss": 5.1836,
|
|
"mean_token_accuracy": 0.2000496044754982,
|
|
"num_tokens": 10676575.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.706257009506226,
|
|
"epoch": 5.051568543188655,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00030391233252733085,
|
|
"loss": 5.1915,
|
|
"mean_token_accuracy": 0.19668924510478974,
|
|
"num_tokens": 10685190.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.740829133987427,
|
|
"epoch": 5.055865921787709,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030358257473447144,
|
|
"loss": 5.2757,
|
|
"mean_token_accuracy": 0.1943893775343895,
|
|
"num_tokens": 10694143.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.744351530075074,
|
|
"epoch": 5.060163300386764,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00030325275452869316,
|
|
"loss": 5.26,
|
|
"mean_token_accuracy": 0.19109616726636885,
|
|
"num_tokens": 10703271.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.782424688339233,
|
|
"epoch": 5.064460678985819,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030292287263019153,
|
|
"loss": 5.3016,
|
|
"mean_token_accuracy": 0.18588456511497498,
|
|
"num_tokens": 10713526.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.809094953536987,
|
|
"epoch": 5.068758057584874,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00030259292975929675,
|
|
"loss": 5.264,
|
|
"mean_token_accuracy": 0.1944745108485222,
|
|
"num_tokens": 10722685.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.753801965713501,
|
|
"epoch": 5.073055436183928,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003022629266364723,
|
|
"loss": 5.2044,
|
|
"mean_token_accuracy": 0.19990399181842805,
|
|
"num_tokens": 10731861.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.684302282333374,
|
|
"epoch": 5.077352814782983,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00030193286398231276,
|
|
"loss": 5.2311,
|
|
"mean_token_accuracy": 0.19561950117349625,
|
|
"num_tokens": 10740291.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.6560767650604244,
|
|
"epoch": 5.081650193382037,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00030160274251754337,
|
|
"loss": 5.1724,
|
|
"mean_token_accuracy": 0.20046201944351197,
|
|
"num_tokens": 10749340.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.756677913665771,
|
|
"epoch": 5.085947571981092,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00030127256296301724,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.19908324331045152,
|
|
"num_tokens": 10758948.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.772564506530761,
|
|
"epoch": 5.090244950580146,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003009423260397148,
|
|
"loss": 5.2686,
|
|
"mean_token_accuracy": 0.1873389646410942,
|
|
"num_tokens": 10768534.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.729564952850342,
|
|
"epoch": 5.094542329179201,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00030061203246874125,
|
|
"loss": 5.1876,
|
|
"mean_token_accuracy": 0.20643589943647384,
|
|
"num_tokens": 10777570.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.733772087097168,
|
|
"epoch": 5.0988397077782555,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030028168297132593,
|
|
"loss": 5.2423,
|
|
"mean_token_accuracy": 0.19843897372484207,
|
|
"num_tokens": 10786952.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.763795661926269,
|
|
"epoch": 5.10313708637731,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002999512782688199,
|
|
"loss": 5.2896,
|
|
"mean_token_accuracy": 0.18885112404823304,
|
|
"num_tokens": 10795401.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.6658929824829105,
|
|
"epoch": 5.1074344649763646,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002996208190826951,
|
|
"loss": 5.1825,
|
|
"mean_token_accuracy": 0.2021891936659813,
|
|
"num_tokens": 10805459.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.733235120773315,
|
|
"epoch": 5.111731843575419,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00029929030613454227,
|
|
"loss": 5.1977,
|
|
"mean_token_accuracy": 0.1947402223944664,
|
|
"num_tokens": 10814635.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.775217437744141,
|
|
"epoch": 5.116029222174474,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002989597401460697,
|
|
"loss": 5.2741,
|
|
"mean_token_accuracy": 0.19305982142686845,
|
|
"num_tokens": 10824256.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.755994129180908,
|
|
"epoch": 5.120326600773528,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029862912183910105,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.19980508685112,
|
|
"num_tokens": 10833919.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.659001350402832,
|
|
"epoch": 5.124623979372583,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00029829845193557496,
|
|
"loss": 5.1367,
|
|
"mean_token_accuracy": 0.20144048631191253,
|
|
"num_tokens": 10842109.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.726154136657715,
|
|
"epoch": 5.128921357971637,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002979677311575421,
|
|
"loss": 5.3294,
|
|
"mean_token_accuracy": 0.19121731519699098,
|
|
"num_tokens": 10851185.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.693465662002564,
|
|
"epoch": 5.133218736570692,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002976369602271646,
|
|
"loss": 5.1662,
|
|
"mean_token_accuracy": 0.2095023438334465,
|
|
"num_tokens": 10859822.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.718322134017944,
|
|
"epoch": 5.137516115169746,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002973061398667138,
|
|
"loss": 5.1845,
|
|
"mean_token_accuracy": 0.1995968461036682,
|
|
"num_tokens": 10868663.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.721281480789185,
|
|
"epoch": 5.141813493768801,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00029697527079856916,
|
|
"loss": 5.1261,
|
|
"mean_token_accuracy": 0.20855843871831894,
|
|
"num_tokens": 10876888.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.7712949275970455,
|
|
"epoch": 5.1461108723678555,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029664435374521665,
|
|
"loss": 5.2672,
|
|
"mean_token_accuracy": 0.19305787235498428,
|
|
"num_tokens": 10885559.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.748762464523315,
|
|
"epoch": 5.15040825096691,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00029631338942924664,
|
|
"loss": 5.2659,
|
|
"mean_token_accuracy": 0.20231701880693437,
|
|
"num_tokens": 10893843.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.704547119140625,
|
|
"epoch": 5.154705629565965,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002959823785733531,
|
|
"loss": 5.195,
|
|
"mean_token_accuracy": 0.19912101477384567,
|
|
"num_tokens": 10903272.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.154705629565965,
|
|
"eval_entropy": 5.56400134058686,
|
|
"eval_loss": 6.024650573730469,
|
|
"eval_mean_token_accuracy": 0.17078697303863796,
|
|
"eval_num_tokens": 10903272.0,
|
|
"eval_runtime": 2.6412,
|
|
"eval_samples_per_second": 1343.704,
|
|
"eval_steps_per_second": 168.105,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.660799551010132,
|
|
"epoch": 5.159003008165019,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002956513219003312,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.19872058629989625,
|
|
"num_tokens": 10912630.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.70833797454834,
|
|
"epoch": 5.163300386764074,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00029532022013307666,
|
|
"loss": 5.1354,
|
|
"mean_token_accuracy": 0.20487826615571975,
|
|
"num_tokens": 10921491.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.681808137893677,
|
|
"epoch": 5.167597765363128,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00029498907399458325,
|
|
"loss": 5.24,
|
|
"mean_token_accuracy": 0.19859227985143663,
|
|
"num_tokens": 10931686.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.719032573699951,
|
|
"epoch": 5.171895143962183,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002946578842079418,
|
|
"loss": 5.3499,
|
|
"mean_token_accuracy": 0.1896876350045204,
|
|
"num_tokens": 10942273.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.767288684844971,
|
|
"epoch": 5.176192522561237,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002943266514963384,
|
|
"loss": 5.3055,
|
|
"mean_token_accuracy": 0.19963981509208678,
|
|
"num_tokens": 10950736.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.777680492401123,
|
|
"epoch": 5.180489901160292,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002939953765830529,
|
|
"loss": 5.229,
|
|
"mean_token_accuracy": 0.1960427224636078,
|
|
"num_tokens": 10959789.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.666653776168824,
|
|
"epoch": 5.184787279759346,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029366406019145735,
|
|
"loss": 5.2058,
|
|
"mean_token_accuracy": 0.19436293989419937,
|
|
"num_tokens": 10968842.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.732349014282226,
|
|
"epoch": 5.189084658358402,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002933327030450143,
|
|
"loss": 5.2211,
|
|
"mean_token_accuracy": 0.1907611221075058,
|
|
"num_tokens": 10977799.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.822662830352783,
|
|
"epoch": 5.193382036957456,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00029300130586727545,
|
|
"loss": 5.2831,
|
|
"mean_token_accuracy": 0.18605872392654418,
|
|
"num_tokens": 10987279.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.7022521018981935,
|
|
"epoch": 5.197679415556511,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00029266986938187943,
|
|
"loss": 5.1975,
|
|
"mean_token_accuracy": 0.2021750330924988,
|
|
"num_tokens": 10996577.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.6809791088104244,
|
|
"epoch": 5.2019767941555655,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002923383943125514,
|
|
"loss": 5.292,
|
|
"mean_token_accuracy": 0.19305624067783356,
|
|
"num_tokens": 11005818.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.764346075057984,
|
|
"epoch": 5.20627417275462,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002920068813831002,
|
|
"loss": 5.1563,
|
|
"mean_token_accuracy": 0.19835299253463745,
|
|
"num_tokens": 11014613.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.698979139328003,
|
|
"epoch": 5.210571551353675,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002916753313174178,
|
|
"loss": 5.2341,
|
|
"mean_token_accuracy": 0.19529004544019699,
|
|
"num_tokens": 11024327.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.6682233810424805,
|
|
"epoch": 5.214868929952729,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0002913437448394768,
|
|
"loss": 5.3325,
|
|
"mean_token_accuracy": 0.20361909866333008,
|
|
"num_tokens": 11034558.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.649209070205688,
|
|
"epoch": 5.219166308551784,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00029101212267332955,
|
|
"loss": 5.1297,
|
|
"mean_token_accuracy": 0.2062030091881752,
|
|
"num_tokens": 11042948.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.76781268119812,
|
|
"epoch": 5.223463687150838,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029068046554310637,
|
|
"loss": 5.2562,
|
|
"mean_token_accuracy": 0.19500191360712052,
|
|
"num_tokens": 11052529.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.7521007537841795,
|
|
"epoch": 5.227761065749893,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002903487741730139,
|
|
"loss": 5.3133,
|
|
"mean_token_accuracy": 0.1894250139594078,
|
|
"num_tokens": 11062464.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.698194980621338,
|
|
"epoch": 5.232058444348947,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00029001704928733354,
|
|
"loss": 5.1686,
|
|
"mean_token_accuracy": 0.2008085146546364,
|
|
"num_tokens": 11070522.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.655770301818848,
|
|
"epoch": 5.236355822948002,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002896852916104198,
|
|
"loss": 5.1342,
|
|
"mean_token_accuracy": 0.2060985431075096,
|
|
"num_tokens": 11080227.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.7664025783538815,
|
|
"epoch": 5.240653201547056,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0002893535018666988,
|
|
"loss": 5.2845,
|
|
"mean_token_accuracy": 0.1933340072631836,
|
|
"num_tokens": 11089040.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.744159936904907,
|
|
"epoch": 5.244950580146111,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00028902168078066674,
|
|
"loss": 5.2901,
|
|
"mean_token_accuracy": 0.1923699676990509,
|
|
"num_tokens": 11098788.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.741655445098877,
|
|
"epoch": 5.2492479587451655,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002886898290768883,
|
|
"loss": 5.1834,
|
|
"mean_token_accuracy": 0.19782494604587555,
|
|
"num_tokens": 11107399.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.6719671249389645,
|
|
"epoch": 5.25354533734422,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000288357947479995,
|
|
"loss": 5.2591,
|
|
"mean_token_accuracy": 0.19018339961767197,
|
|
"num_tokens": 11116833.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.712065935134888,
|
|
"epoch": 5.257842715943275,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028802603671468347,
|
|
"loss": 5.2427,
|
|
"mean_token_accuracy": 0.19288953989744187,
|
|
"num_tokens": 11126027.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.793352937698364,
|
|
"epoch": 5.262140094542329,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00028769409750571413,
|
|
"loss": 5.297,
|
|
"mean_token_accuracy": 0.18685111850500108,
|
|
"num_tokens": 11135873.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.655130195617676,
|
|
"epoch": 5.266437473141384,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00028736213057790975,
|
|
"loss": 5.0623,
|
|
"mean_token_accuracy": 0.2107608959078789,
|
|
"num_tokens": 11144470.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.658469724655151,
|
|
"epoch": 5.270734851740438,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002870301366561533,
|
|
"loss": 5.2174,
|
|
"mean_token_accuracy": 0.20162947326898575,
|
|
"num_tokens": 11153531.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.711200857162476,
|
|
"epoch": 5.275032230339493,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002866981164653867,
|
|
"loss": 5.255,
|
|
"mean_token_accuracy": 0.18732093423604965,
|
|
"num_tokens": 11163365.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.709438180923462,
|
|
"epoch": 5.279329608938547,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002863660707306095,
|
|
"loss": 5.1607,
|
|
"mean_token_accuracy": 0.19922513514757156,
|
|
"num_tokens": 11171907.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.661566734313965,
|
|
"epoch": 5.283626987537602,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00028603400017687675,
|
|
"loss": 5.2422,
|
|
"mean_token_accuracy": 0.19477804899215698,
|
|
"num_tokens": 11180804.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.79567403793335,
|
|
"epoch": 5.287924366136656,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00028570190552929794,
|
|
"loss": 5.3035,
|
|
"mean_token_accuracy": 0.18904237151145936,
|
|
"num_tokens": 11189762.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.686970329284668,
|
|
"epoch": 5.292221744735711,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000285369787513035,
|
|
"loss": 5.1997,
|
|
"mean_token_accuracy": 0.19904248118400575,
|
|
"num_tokens": 11197983.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.6867420196533205,
|
|
"epoch": 5.2965191233347655,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00028503764685330077,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.20757964253425598,
|
|
"num_tokens": 11206286.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.718180322647095,
|
|
"epoch": 5.30081650193382,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00028470548427535794,
|
|
"loss": 5.2678,
|
|
"mean_token_accuracy": 0.19468450099229812,
|
|
"num_tokens": 11216444.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.662199354171753,
|
|
"epoch": 5.305113880532875,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028437330050451654,
|
|
"loss": 5.1874,
|
|
"mean_token_accuracy": 0.20235961377620698,
|
|
"num_tokens": 11225378.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.717228269577026,
|
|
"epoch": 5.309411259131929,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002840410962661334,
|
|
"loss": 5.3219,
|
|
"mean_token_accuracy": 0.1865878000855446,
|
|
"num_tokens": 11234352.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.7377190589904785,
|
|
"epoch": 5.313708637730985,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002837088722856098,
|
|
"loss": 5.2215,
|
|
"mean_token_accuracy": 0.2032409206032753,
|
|
"num_tokens": 11243350.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.73460168838501,
|
|
"epoch": 5.318006016330038,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00028337662928838996,
|
|
"loss": 5.2402,
|
|
"mean_token_accuracy": 0.19373822212219238,
|
|
"num_tokens": 11252746.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.690977954864502,
|
|
"epoch": 5.322303394929094,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00028304436799995986,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.19862150996923447,
|
|
"num_tokens": 11262905.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.691837644577026,
|
|
"epoch": 5.326600773528148,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00028271208914584534,
|
|
"loss": 5.213,
|
|
"mean_token_accuracy": 0.20114982426166533,
|
|
"num_tokens": 11270925.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.718298625946045,
|
|
"epoch": 5.330898152127203,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00028237979345161065,
|
|
"loss": 5.2222,
|
|
"mean_token_accuracy": 0.19509415775537492,
|
|
"num_tokens": 11279689.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.68168272972107,
|
|
"epoch": 5.335195530726257,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002820474816428568,
|
|
"loss": 5.1673,
|
|
"mean_token_accuracy": 0.20950797349214553,
|
|
"num_tokens": 11288783.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.569508981704712,
|
|
"epoch": 5.339492909325312,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002817151544452198,
|
|
"loss": 5.1086,
|
|
"mean_token_accuracy": 0.21557158827781678,
|
|
"num_tokens": 11297903.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.690628242492676,
|
|
"epoch": 5.343790287924366,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00028138281258436947,
|
|
"loss": 5.1786,
|
|
"mean_token_accuracy": 0.20744261443614959,
|
|
"num_tokens": 11306591.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.709564924240112,
|
|
"epoch": 5.348087666523421,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002810504567860078,
|
|
"loss": 5.2542,
|
|
"mean_token_accuracy": 0.19505104869604112,
|
|
"num_tokens": 11316772.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.701749086380005,
|
|
"epoch": 5.3523850451224755,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002807180877758667,
|
|
"loss": 5.2344,
|
|
"mean_token_accuracy": 0.19660228341817856,
|
|
"num_tokens": 11326842.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.641543292999268,
|
|
"epoch": 5.35668242372153,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00028038570627970754,
|
|
"loss": 5.2621,
|
|
"mean_token_accuracy": 0.19944286197423935,
|
|
"num_tokens": 11335765.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.743082809448242,
|
|
"epoch": 5.360979802320585,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002800533130233184,
|
|
"loss": 5.316,
|
|
"mean_token_accuracy": 0.19234861731529235,
|
|
"num_tokens": 11345140.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.720323324203491,
|
|
"epoch": 5.365277180919639,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.0002797209087325135,
|
|
"loss": 5.1782,
|
|
"mean_token_accuracy": 0.19889781177043914,
|
|
"num_tokens": 11354585.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.772159671783447,
|
|
"epoch": 5.369574559518694,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00027938849413313083,
|
|
"loss": 5.3605,
|
|
"mean_token_accuracy": 0.1926327496767044,
|
|
"num_tokens": 11364606.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.6875832080841064,
|
|
"epoch": 5.373871938117748,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000279056069951031,
|
|
"loss": 5.2797,
|
|
"mean_token_accuracy": 0.19274755865335463,
|
|
"num_tokens": 11374399.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.692865896224975,
|
|
"epoch": 5.378169316716803,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00027872363691209564,
|
|
"loss": 5.1416,
|
|
"mean_token_accuracy": 0.2035885214805603,
|
|
"num_tokens": 11383040.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.692754030227661,
|
|
"epoch": 5.382466695315857,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002783911957422256,
|
|
"loss": 5.2981,
|
|
"mean_token_accuracy": 0.1915803074836731,
|
|
"num_tokens": 11392088.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.694613981246948,
|
|
"epoch": 5.386764073914912,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002780587471673394,
|
|
"loss": 5.2067,
|
|
"mean_token_accuracy": 0.19618823230266572,
|
|
"num_tokens": 11401400.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.6945232391357425,
|
|
"epoch": 5.391061452513966,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00027772629191337206,
|
|
"loss": 5.2226,
|
|
"mean_token_accuracy": 0.20668426901102066,
|
|
"num_tokens": 11410205.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.734327507019043,
|
|
"epoch": 5.395358831113021,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00027739383070627283,
|
|
"loss": 5.2263,
|
|
"mean_token_accuracy": 0.2047954887151718,
|
|
"num_tokens": 11419208.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.618894815444946,
|
|
"epoch": 5.3996562097120755,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002770613642720041,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.21685144901275635,
|
|
"num_tokens": 11427386.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.5679309368133545,
|
|
"epoch": 5.40395358831113,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00027672889333653984,
|
|
"loss": 5.2028,
|
|
"mean_token_accuracy": 0.20548076182603836,
|
|
"num_tokens": 11436203.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.660821199417114,
|
|
"epoch": 5.408250966910185,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002763964186258635,
|
|
"loss": 5.2839,
|
|
"mean_token_accuracy": 0.1874884083867073,
|
|
"num_tokens": 11445682.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.7353109359741214,
|
|
"epoch": 5.412548345509239,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002760639408659671,
|
|
"loss": 5.1929,
|
|
"mean_token_accuracy": 0.2019863471388817,
|
|
"num_tokens": 11454622.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.667498064041138,
|
|
"epoch": 5.416845724108294,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002757314607828489,
|
|
"loss": 5.1663,
|
|
"mean_token_accuracy": 0.20001015812158585,
|
|
"num_tokens": 11463539.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.699609661102295,
|
|
"epoch": 5.421143102707348,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00027539897910251293,
|
|
"loss": 5.1647,
|
|
"mean_token_accuracy": 0.20516479313373565,
|
|
"num_tokens": 11472491.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.617361068725586,
|
|
"epoch": 5.425440481306403,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00027506649655096595,
|
|
"loss": 5.2203,
|
|
"mean_token_accuracy": 0.19524253010749817,
|
|
"num_tokens": 11480899.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.658642244338989,
|
|
"epoch": 5.429737859905457,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002747340138542171,
|
|
"loss": 5.2185,
|
|
"mean_token_accuracy": 0.19106969982385635,
|
|
"num_tokens": 11490525.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.664838409423828,
|
|
"epoch": 5.434035238504512,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002744015317382757,
|
|
"loss": 5.2113,
|
|
"mean_token_accuracy": 0.1966550350189209,
|
|
"num_tokens": 11499589.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.756063652038574,
|
|
"epoch": 5.438332617103566,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002740690509291498,
|
|
"loss": 5.288,
|
|
"mean_token_accuracy": 0.19049128144979477,
|
|
"num_tokens": 11508425.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.728606796264648,
|
|
"epoch": 5.442629995702621,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002737365721528445,
|
|
"loss": 5.2781,
|
|
"mean_token_accuracy": 0.19733783304691316,
|
|
"num_tokens": 11517723.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.648084402084351,
|
|
"epoch": 5.446927374301676,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002734040961353607,
|
|
"loss": 5.1438,
|
|
"mean_token_accuracy": 0.2037380740046501,
|
|
"num_tokens": 11525882.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.651482677459716,
|
|
"epoch": 5.451224752900731,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000273071623602693,
|
|
"loss": 5.317,
|
|
"mean_token_accuracy": 0.18950090408325196,
|
|
"num_tokens": 11535646.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.712294054031372,
|
|
"epoch": 5.4555221314997855,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00027273915528082865,
|
|
"loss": 5.1582,
|
|
"mean_token_accuracy": 0.2003449335694313,
|
|
"num_tokens": 11545413.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.630814695358277,
|
|
"epoch": 5.45981951009884,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002724066918957455,
|
|
"loss": 5.1107,
|
|
"mean_token_accuracy": 0.21045234501361848,
|
|
"num_tokens": 11554680.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.782881832122802,
|
|
"epoch": 5.464116888697895,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002720742341734107,
|
|
"loss": 5.3007,
|
|
"mean_token_accuracy": 0.1973700001835823,
|
|
"num_tokens": 11564607.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.657800102233887,
|
|
"epoch": 5.468414267296949,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027174178283977904,
|
|
"loss": 5.2661,
|
|
"mean_token_accuracy": 0.19993091821670533,
|
|
"num_tokens": 11574097.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.615719795227051,
|
|
"epoch": 5.472711645896004,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027140933862079136,
|
|
"loss": 5.2054,
|
|
"mean_token_accuracy": 0.20218315720558167,
|
|
"num_tokens": 11582561.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.602358055114746,
|
|
"epoch": 5.477009024495058,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000271076902242373,
|
|
"loss": 5.2642,
|
|
"mean_token_accuracy": 0.18935706913471223,
|
|
"num_tokens": 11591747.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.712296009063721,
|
|
"epoch": 5.481306403094113,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000270744474430432,
|
|
"loss": 5.2645,
|
|
"mean_token_accuracy": 0.19682457000017167,
|
|
"num_tokens": 11599928.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.667343807220459,
|
|
"epoch": 5.485603781693167,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000270412055910858,
|
|
"loss": 5.1681,
|
|
"mean_token_accuracy": 0.2059781402349472,
|
|
"num_tokens": 11607912.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.675365829467774,
|
|
"epoch": 5.489901160292222,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002700796474095201,
|
|
"loss": 5.3032,
|
|
"mean_token_accuracy": 0.18909088969230653,
|
|
"num_tokens": 11617407.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.664082717895508,
|
|
"epoch": 5.494198538891276,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002697472496522656,
|
|
"loss": 5.1439,
|
|
"mean_token_accuracy": 0.2086104929447174,
|
|
"num_tokens": 11626280.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.730630779266358,
|
|
"epoch": 5.498495917490331,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002694148633649184,
|
|
"loss": 5.2026,
|
|
"mean_token_accuracy": 0.20186474323272705,
|
|
"num_tokens": 11634959.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.693294191360474,
|
|
"epoch": 5.5027932960893855,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002690824892732772,
|
|
"loss": 5.2404,
|
|
"mean_token_accuracy": 0.189927314221859,
|
|
"num_tokens": 11644044.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.662776231765747,
|
|
"epoch": 5.50709067468844,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002687501281031142,
|
|
"loss": 5.187,
|
|
"mean_token_accuracy": 0.2064729481935501,
|
|
"num_tokens": 11653622.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.796233654022217,
|
|
"epoch": 5.511388053287495,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0002684177805801734,
|
|
"loss": 5.3135,
|
|
"mean_token_accuracy": 0.18745039999485016,
|
|
"num_tokens": 11663716.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.776729536056519,
|
|
"epoch": 5.515685431886549,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00026808544743016886,
|
|
"loss": 5.3481,
|
|
"mean_token_accuracy": 0.1895677700638771,
|
|
"num_tokens": 11672769.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.754831314086914,
|
|
"epoch": 5.519982810485604,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002677531293787835,
|
|
"loss": 5.2298,
|
|
"mean_token_accuracy": 0.2020084008574486,
|
|
"num_tokens": 11682492.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.674351835250855,
|
|
"epoch": 5.524280189084658,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000267420827151667,
|
|
"loss": 5.2258,
|
|
"mean_token_accuracy": 0.1978085607290268,
|
|
"num_tokens": 11690738.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.592818784713745,
|
|
"epoch": 5.528577567683713,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002670885414744347,
|
|
"loss": 5.1353,
|
|
"mean_token_accuracy": 0.20714132040739058,
|
|
"num_tokens": 11699925.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.6374414443969725,
|
|
"epoch": 5.532874946282767,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002667562730726655,
|
|
"loss": 5.2102,
|
|
"mean_token_accuracy": 0.198835526406765,
|
|
"num_tokens": 11709291.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.683079099655151,
|
|
"epoch": 5.537172324881822,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00026642402267190095,
|
|
"loss": 5.2317,
|
|
"mean_token_accuracy": 0.20534752160310746,
|
|
"num_tokens": 11717958.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.6939221858978275,
|
|
"epoch": 5.541469703480876,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00026609179099764313,
|
|
"loss": 5.2337,
|
|
"mean_token_accuracy": 0.19280954003334044,
|
|
"num_tokens": 11726776.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.667202711105347,
|
|
"epoch": 5.545767082079931,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00026575957877535323,
|
|
"loss": 5.2261,
|
|
"mean_token_accuracy": 0.2039433091878891,
|
|
"num_tokens": 11735913.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.713533496856689,
|
|
"epoch": 5.5500644606789855,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00026542738673044985,
|
|
"loss": 5.2331,
|
|
"mean_token_accuracy": 0.1948437809944153,
|
|
"num_tokens": 11744728.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.7503269672393795,
|
|
"epoch": 5.55436183927804,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0002650952155883077,
|
|
"loss": 5.2641,
|
|
"mean_token_accuracy": 0.1911979630589485,
|
|
"num_tokens": 11753440.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.668906402587891,
|
|
"epoch": 5.558659217877095,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002647630660742559,
|
|
"loss": 5.1864,
|
|
"mean_token_accuracy": 0.21024998128414155,
|
|
"num_tokens": 11762061.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.707573366165161,
|
|
"epoch": 5.56295659647615,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000264430938913576,
|
|
"loss": 5.2661,
|
|
"mean_token_accuracy": 0.186495241522789,
|
|
"num_tokens": 11771421.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.613175392150879,
|
|
"epoch": 5.567253975075204,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00026409883483150123,
|
|
"loss": 5.1574,
|
|
"mean_token_accuracy": 0.2117705374956131,
|
|
"num_tokens": 11781062.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.731540060043335,
|
|
"epoch": 5.571551353674259,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000263766754553214,
|
|
"loss": 5.2755,
|
|
"mean_token_accuracy": 0.19592178761959075,
|
|
"num_tokens": 11790188.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.702342128753662,
|
|
"epoch": 5.575848732273313,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002634346988038448,
|
|
"loss": 5.2655,
|
|
"mean_token_accuracy": 0.19424394220113755,
|
|
"num_tokens": 11799093.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.71532187461853,
|
|
"epoch": 5.580146110872368,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026310266830847093,
|
|
"loss": 5.1881,
|
|
"mean_token_accuracy": 0.20259663462638855,
|
|
"num_tokens": 11808357.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.635984754562378,
|
|
"epoch": 5.584443489471423,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026277066379211406,
|
|
"loss": 5.2266,
|
|
"mean_token_accuracy": 0.1998278334736824,
|
|
"num_tokens": 11817229.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.584443489471423,
|
|
"eval_entropy": 5.546964949852711,
|
|
"eval_loss": 6.005390644073486,
|
|
"eval_mean_token_accuracy": 0.17231312395820209,
|
|
"eval_num_tokens": 11817229.0,
|
|
"eval_runtime": 2.6403,
|
|
"eval_samples_per_second": 1344.162,
|
|
"eval_steps_per_second": 168.162,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.692148351669312,
|
|
"epoch": 5.588740868070477,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002624386859797396,
|
|
"loss": 5.2682,
|
|
"mean_token_accuracy": 0.1935577303171158,
|
|
"num_tokens": 11826952.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.695202016830445,
|
|
"epoch": 5.593038246669532,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00026210673559625406,
|
|
"loss": 5.2364,
|
|
"mean_token_accuracy": 0.19843607544898986,
|
|
"num_tokens": 11836411.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.757038211822509,
|
|
"epoch": 5.597335625268586,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002617748133665047,
|
|
"loss": 5.305,
|
|
"mean_token_accuracy": 0.196697635948658,
|
|
"num_tokens": 11846140.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.739417600631714,
|
|
"epoch": 5.601633003867641,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0002614429200152768,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.20118748843669892,
|
|
"num_tokens": 11855057.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.705063962936402,
|
|
"epoch": 5.6059303824666955,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000261111056267293,
|
|
"loss": 5.1842,
|
|
"mean_token_accuracy": 0.2024633675813675,
|
|
"num_tokens": 11863257.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.717278051376343,
|
|
"epoch": 5.61022776106575,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00026077922284721084,
|
|
"loss": 5.3335,
|
|
"mean_token_accuracy": 0.1914536699652672,
|
|
"num_tokens": 11872706.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.6277566909790036,
|
|
"epoch": 5.614525139664805,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026044742047962206,
|
|
"loss": 5.2335,
|
|
"mean_token_accuracy": 0.19924333840608596,
|
|
"num_tokens": 11882692.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.7170874118804935,
|
|
"epoch": 5.618822518263859,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00026011564988905023,
|
|
"loss": 5.1875,
|
|
"mean_token_accuracy": 0.20180433988571167,
|
|
"num_tokens": 11893152.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.795899248123169,
|
|
"epoch": 5.623119896862914,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002597839117999499,
|
|
"loss": 5.2503,
|
|
"mean_token_accuracy": 0.1924741879105568,
|
|
"num_tokens": 11902193.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.65684700012207,
|
|
"epoch": 5.627417275461968,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002594522069367044,
|
|
"loss": 5.2276,
|
|
"mean_token_accuracy": 0.20079947412014007,
|
|
"num_tokens": 11910909.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.648374319076538,
|
|
"epoch": 5.631714654061023,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002591205360236245,
|
|
"loss": 5.2386,
|
|
"mean_token_accuracy": 0.19885893017053605,
|
|
"num_tokens": 11919518.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.700269269943237,
|
|
"epoch": 5.636012032660077,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000258788899784947,
|
|
"loss": 5.2304,
|
|
"mean_token_accuracy": 0.19707950800657273,
|
|
"num_tokens": 11928735.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.685926914215088,
|
|
"epoch": 5.640309411259132,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00025845729894483283,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.20316193699836732,
|
|
"num_tokens": 11937266.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.609309673309326,
|
|
"epoch": 5.644606789858186,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002581257342273657,
|
|
"loss": 5.1786,
|
|
"mean_token_accuracy": 0.19774096459150314,
|
|
"num_tokens": 11946583.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.72201247215271,
|
|
"epoch": 5.648904168457241,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0002577942063565504,
|
|
"loss": 5.2758,
|
|
"mean_token_accuracy": 0.1922171413898468,
|
|
"num_tokens": 11955883.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.771066522598266,
|
|
"epoch": 5.6532015470562955,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002574627160563114,
|
|
"loss": 5.2022,
|
|
"mean_token_accuracy": 0.2036671057343483,
|
|
"num_tokens": 11965340.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.592436742782593,
|
|
"epoch": 5.65749892565535,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002571312640504909,
|
|
"loss": 5.1423,
|
|
"mean_token_accuracy": 0.20673393905162812,
|
|
"num_tokens": 11974389.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.708108949661255,
|
|
"epoch": 5.661796304254405,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002567998510628476,
|
|
"loss": 5.3161,
|
|
"mean_token_accuracy": 0.1902399927377701,
|
|
"num_tokens": 11983872.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.734968614578247,
|
|
"epoch": 5.666093682853459,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025646847781705506,
|
|
"loss": 5.1825,
|
|
"mean_token_accuracy": 0.20284292697906495,
|
|
"num_tokens": 11992318.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.687312746047974,
|
|
"epoch": 5.670391061452514,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002561371450367,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.19871245175600052,
|
|
"num_tokens": 12001926.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.683475494384766,
|
|
"epoch": 5.674688440051568,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00025580585344528076,
|
|
"loss": 5.3211,
|
|
"mean_token_accuracy": 0.19579958617687226,
|
|
"num_tokens": 12011505.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.707354593276977,
|
|
"epoch": 5.678985818650623,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002554746037662058,
|
|
"loss": 5.1802,
|
|
"mean_token_accuracy": 0.1996296763420105,
|
|
"num_tokens": 12020969.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.743630313873291,
|
|
"epoch": 5.683283197249677,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002551433967227919,
|
|
"loss": 5.2413,
|
|
"mean_token_accuracy": 0.1997652307152748,
|
|
"num_tokens": 12029786.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.782976675033569,
|
|
"epoch": 5.687580575848732,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000254812233038263,
|
|
"loss": 5.3235,
|
|
"mean_token_accuracy": 0.18913015276193618,
|
|
"num_tokens": 12039081.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.699254703521729,
|
|
"epoch": 5.691877954447786,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00025448111343574813,
|
|
"loss": 5.2517,
|
|
"mean_token_accuracy": 0.1981458008289337,
|
|
"num_tokens": 12049129.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.679174089431763,
|
|
"epoch": 5.696175333046842,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002541500386382802,
|
|
"loss": 5.0932,
|
|
"mean_token_accuracy": 0.21781291216611862,
|
|
"num_tokens": 12058243.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.627520275115967,
|
|
"epoch": 5.7004727116458955,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00025381900936879433,
|
|
"loss": 5.0888,
|
|
"mean_token_accuracy": 0.21505894809961318,
|
|
"num_tokens": 12067468.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.6727149963378904,
|
|
"epoch": 5.704770090244951,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002534880263501259,
|
|
"loss": 5.1999,
|
|
"mean_token_accuracy": 0.19918537586927415,
|
|
"num_tokens": 12076090.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.714193964004517,
|
|
"epoch": 5.7090674688440055,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002531570903050097,
|
|
"loss": 5.2961,
|
|
"mean_token_accuracy": 0.19609350860118865,
|
|
"num_tokens": 12085614.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.655387353897095,
|
|
"epoch": 5.71336484744306,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002528262019560776,
|
|
"loss": 5.2042,
|
|
"mean_token_accuracy": 0.19804851710796356,
|
|
"num_tokens": 12094291.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.730686140060425,
|
|
"epoch": 5.717662226042115,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002524953620258579,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.19659732282161713,
|
|
"num_tokens": 12102478.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.745828008651733,
|
|
"epoch": 5.721959604641169,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002521645712367724,
|
|
"loss": 5.3547,
|
|
"mean_token_accuracy": 0.18684251457452775,
|
|
"num_tokens": 12112554.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.739840269088745,
|
|
"epoch": 5.726256983240224,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00025183383031113606,
|
|
"loss": 5.2946,
|
|
"mean_token_accuracy": 0.19164000898599626,
|
|
"num_tokens": 12121685.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.754913187026977,
|
|
"epoch": 5.730554361839278,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00025150313997115476,
|
|
"loss": 5.2957,
|
|
"mean_token_accuracy": 0.19254420697689056,
|
|
"num_tokens": 12131980.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.715495586395264,
|
|
"epoch": 5.734851740438333,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.0002511725009389244,
|
|
"loss": 5.245,
|
|
"mean_token_accuracy": 0.19942157864570617,
|
|
"num_tokens": 12141457.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.8150887966156,
|
|
"epoch": 5.739149119037387,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000250841913936428,
|
|
"loss": 5.3148,
|
|
"mean_token_accuracy": 0.19357635974884033,
|
|
"num_tokens": 12149803.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.777203750610352,
|
|
"epoch": 5.743446497636442,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002505113796855357,
|
|
"loss": 5.181,
|
|
"mean_token_accuracy": 0.20085709542036057,
|
|
"num_tokens": 12157904.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.6501928806304935,
|
|
"epoch": 5.747743876235496,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00025018089890800225,
|
|
"loss": 5.2107,
|
|
"mean_token_accuracy": 0.198909392952919,
|
|
"num_tokens": 12166540.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.602605295181275,
|
|
"epoch": 5.752041254834551,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024985047232546544,
|
|
"loss": 5.2277,
|
|
"mean_token_accuracy": 0.201987524330616,
|
|
"num_tokens": 12175345.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.7576159000396725,
|
|
"epoch": 5.7563386334336055,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024952010065944485,
|
|
"loss": 5.2684,
|
|
"mean_token_accuracy": 0.19707789570093154,
|
|
"num_tokens": 12184971.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.693032264709473,
|
|
"epoch": 5.76063601203266,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002491897846313402,
|
|
"loss": 5.1674,
|
|
"mean_token_accuracy": 0.21217764168977737,
|
|
"num_tokens": 12193643.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.6620824337005615,
|
|
"epoch": 5.764933390631715,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002488595249624297,
|
|
"loss": 5.3048,
|
|
"mean_token_accuracy": 0.18859235644340516,
|
|
"num_tokens": 12203073.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.678703832626343,
|
|
"epoch": 5.769230769230769,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00024852932237386837,
|
|
"loss": 5.1322,
|
|
"mean_token_accuracy": 0.20321830958127976,
|
|
"num_tokens": 12211708.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.636871290206909,
|
|
"epoch": 5.773528147829824,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024819917758668673,
|
|
"loss": 5.1674,
|
|
"mean_token_accuracy": 0.2026461109519005,
|
|
"num_tokens": 12219980.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.748016119003296,
|
|
"epoch": 5.777825526428878,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024786909132178906,
|
|
"loss": 5.3472,
|
|
"mean_token_accuracy": 0.182881635427475,
|
|
"num_tokens": 12229523.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.698806381225586,
|
|
"epoch": 5.782122905027933,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00024753906429995194,
|
|
"loss": 5.1545,
|
|
"mean_token_accuracy": 0.2060826987028122,
|
|
"num_tokens": 12238275.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.588642215728759,
|
|
"epoch": 5.786420283626987,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002472090972418222,
|
|
"loss": 5.2002,
|
|
"mean_token_accuracy": 0.20266135931015014,
|
|
"num_tokens": 12247569.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.662156057357788,
|
|
"epoch": 5.790717662226042,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002468791908679163,
|
|
"loss": 5.3031,
|
|
"mean_token_accuracy": 0.19203459918498994,
|
|
"num_tokens": 12256923.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.694774341583252,
|
|
"epoch": 5.795015040825096,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002465493458986175,
|
|
"loss": 5.1856,
|
|
"mean_token_accuracy": 0.2124566078186035,
|
|
"num_tokens": 12265607.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.722589731216431,
|
|
"epoch": 5.799312419424151,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024621956305417587,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.19813698083162307,
|
|
"num_tokens": 12274348.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.6771753311157225,
|
|
"epoch": 5.8036097980232055,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000245889843054705,
|
|
"loss": 5.2392,
|
|
"mean_token_accuracy": 0.20069831013679504,
|
|
"num_tokens": 12283072.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.641410779953003,
|
|
"epoch": 5.80790717662226,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00024556018662018163,
|
|
"loss": 5.2364,
|
|
"mean_token_accuracy": 0.20008289515972139,
|
|
"num_tokens": 12292390.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.707869005203247,
|
|
"epoch": 5.812204555221315,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00024523059447044377,
|
|
"loss": 5.2581,
|
|
"mean_token_accuracy": 0.18992670774459838,
|
|
"num_tokens": 12301394.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.748634767532349,
|
|
"epoch": 5.816501933820369,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002449010673251887,
|
|
"loss": 5.2386,
|
|
"mean_token_accuracy": 0.1998512014746666,
|
|
"num_tokens": 12310926.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.758015537261963,
|
|
"epoch": 5.820799312419425,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0002445716059039723,
|
|
"loss": 5.2581,
|
|
"mean_token_accuracy": 0.19584919065237044,
|
|
"num_tokens": 12319797.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.702878952026367,
|
|
"epoch": 5.825096691018478,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024424221092620644,
|
|
"loss": 5.3138,
|
|
"mean_token_accuracy": 0.1885065495967865,
|
|
"num_tokens": 12329776.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.6744529724121096,
|
|
"epoch": 5.829394069617534,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024391288311115822,
|
|
"loss": 5.2802,
|
|
"mean_token_accuracy": 0.19108166694641113,
|
|
"num_tokens": 12339679.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.695830249786377,
|
|
"epoch": 5.833691448216588,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002435836231779478,
|
|
"loss": 5.2002,
|
|
"mean_token_accuracy": 0.2092976227402687,
|
|
"num_tokens": 12347647.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.720023250579834,
|
|
"epoch": 5.837988826815643,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00024325443184554724,
|
|
"loss": 5.2264,
|
|
"mean_token_accuracy": 0.19989611506462096,
|
|
"num_tokens": 12356567.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.709583473205567,
|
|
"epoch": 5.842286205414697,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024292530983277904,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.19673217236995696,
|
|
"num_tokens": 12365434.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.6818156242370605,
|
|
"epoch": 5.846583584013752,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00024259625785831408,
|
|
"loss": 5.2805,
|
|
"mean_token_accuracy": 0.19330639243125916,
|
|
"num_tokens": 12373600.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.754310750961304,
|
|
"epoch": 5.850880962612806,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00024226727664067023,
|
|
"loss": 5.2698,
|
|
"mean_token_accuracy": 0.1963975101709366,
|
|
"num_tokens": 12382979.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.760396099090576,
|
|
"epoch": 5.855178341211861,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024193836689821109,
|
|
"loss": 5.2701,
|
|
"mean_token_accuracy": 0.18660116493701934,
|
|
"num_tokens": 12392142.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.7081231594085695,
|
|
"epoch": 5.8594757198109155,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002416095293491439,
|
|
"loss": 5.1896,
|
|
"mean_token_accuracy": 0.20407740473747255,
|
|
"num_tokens": 12401366.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.644209623336792,
|
|
"epoch": 5.86377309840997,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002412807647115186,
|
|
"loss": 5.1395,
|
|
"mean_token_accuracy": 0.20752381831407546,
|
|
"num_tokens": 12410081.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.643818140029907,
|
|
"epoch": 5.868070477009025,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024095207370322574,
|
|
"loss": 5.1217,
|
|
"mean_token_accuracy": 0.21109537780284882,
|
|
"num_tokens": 12419662.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.65108814239502,
|
|
"epoch": 5.872367855608079,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00024062345704199507,
|
|
"loss": 5.2801,
|
|
"mean_token_accuracy": 0.1933308631181717,
|
|
"num_tokens": 12429202.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.700621747970581,
|
|
"epoch": 5.876665234207134,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00024029491544539405,
|
|
"loss": 5.2072,
|
|
"mean_token_accuracy": 0.20399122387170793,
|
|
"num_tokens": 12438060.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.632294464111328,
|
|
"epoch": 5.880962612806188,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023996644963082616,
|
|
"loss": 5.154,
|
|
"mean_token_accuracy": 0.20825054347515107,
|
|
"num_tokens": 12446897.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.678905057907104,
|
|
"epoch": 5.885259991405243,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00023963806031552948,
|
|
"loss": 5.2797,
|
|
"mean_token_accuracy": 0.18955134451389313,
|
|
"num_tokens": 12456273.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.6969109058380125,
|
|
"epoch": 5.889557370004297,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023930974821657504,
|
|
"loss": 5.2444,
|
|
"mean_token_accuracy": 0.19643399715423585,
|
|
"num_tokens": 12465183.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.739516305923462,
|
|
"epoch": 5.893854748603352,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023898151405086533,
|
|
"loss": 5.312,
|
|
"mean_token_accuracy": 0.19061943143606186,
|
|
"num_tokens": 12474248.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.750906896591187,
|
|
"epoch": 5.8981521272024064,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00023865335853513232,
|
|
"loss": 5.25,
|
|
"mean_token_accuracy": 0.2043888747692108,
|
|
"num_tokens": 12483383.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.700028896331787,
|
|
"epoch": 5.902449505801461,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00023832528238593677,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.20670687556266784,
|
|
"num_tokens": 12491707.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.684899950027466,
|
|
"epoch": 5.9067468844005155,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023799728631966556,
|
|
"loss": 5.2245,
|
|
"mean_token_accuracy": 0.1959487333893776,
|
|
"num_tokens": 12501277.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.749382638931275,
|
|
"epoch": 5.91104426299957,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002376693710525313,
|
|
"loss": 5.2168,
|
|
"mean_token_accuracy": 0.19708582758903503,
|
|
"num_tokens": 12509655.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.696631050109863,
|
|
"epoch": 5.915341641598625,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023734153730056967,
|
|
"loss": 5.1943,
|
|
"mean_token_accuracy": 0.20529661625623702,
|
|
"num_tokens": 12518408.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.742646598815918,
|
|
"epoch": 5.919639020197679,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00023701378577963873,
|
|
"loss": 5.2497,
|
|
"mean_token_accuracy": 0.20038098096847534,
|
|
"num_tokens": 12528502.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.65311017036438,
|
|
"epoch": 5.923936398796734,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002366861172054166,
|
|
"loss": 5.208,
|
|
"mean_token_accuracy": 0.19989775419235228,
|
|
"num_tokens": 12537803.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.732144546508789,
|
|
"epoch": 5.928233777395788,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00023635853229340054,
|
|
"loss": 5.2103,
|
|
"mean_token_accuracy": 0.19834264367818832,
|
|
"num_tokens": 12547432.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.753106164932251,
|
|
"epoch": 5.932531155994843,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023603103175890512,
|
|
"loss": 5.2897,
|
|
"mean_token_accuracy": 0.19138555675745011,
|
|
"num_tokens": 12557630.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.664933681488037,
|
|
"epoch": 5.936828534593897,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00023570361631706062,
|
|
"loss": 5.2127,
|
|
"mean_token_accuracy": 0.19631093442440034,
|
|
"num_tokens": 12566168.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.694322967529297,
|
|
"epoch": 5.941125913192952,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023537628668281142,
|
|
"loss": 5.2461,
|
|
"mean_token_accuracy": 0.19300967305898667,
|
|
"num_tokens": 12575523.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.700940942764282,
|
|
"epoch": 5.945423291792007,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00023504904357091468,
|
|
"loss": 5.1313,
|
|
"mean_token_accuracy": 0.21341818571090698,
|
|
"num_tokens": 12583598.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.56794285774231,
|
|
"epoch": 5.949720670391061,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002347218876959384,
|
|
"loss": 5.1775,
|
|
"mean_token_accuracy": 0.1960050344467163,
|
|
"num_tokens": 12592627.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.595591878890991,
|
|
"epoch": 5.954018048990116,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002343948197722604,
|
|
"loss": 5.2051,
|
|
"mean_token_accuracy": 0.1980335682630539,
|
|
"num_tokens": 12601566.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.6438469886779785,
|
|
"epoch": 5.958315427589171,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023406784051406638,
|
|
"loss": 5.2597,
|
|
"mean_token_accuracy": 0.19859346747398376,
|
|
"num_tokens": 12610526.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.709021186828613,
|
|
"epoch": 5.9626128061882255,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00023374095063534816,
|
|
"loss": 5.2174,
|
|
"mean_token_accuracy": 0.19308471381664277,
|
|
"num_tokens": 12618892.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.756448650360108,
|
|
"epoch": 5.96691018478728,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00023341415084990276,
|
|
"loss": 5.2627,
|
|
"mean_token_accuracy": 0.1964446485042572,
|
|
"num_tokens": 12627905.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.7023333549499515,
|
|
"epoch": 5.971207563386335,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00023308744187132996,
|
|
"loss": 5.2419,
|
|
"mean_token_accuracy": 0.19438775777816772,
|
|
"num_tokens": 12636593.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.679685974121094,
|
|
"epoch": 5.975504941985389,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023276082441303197,
|
|
"loss": 5.1962,
|
|
"mean_token_accuracy": 0.20403438955545425,
|
|
"num_tokens": 12646055.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.7241720199584964,
|
|
"epoch": 5.979802320584444,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00023243429918821056,
|
|
"loss": 5.2555,
|
|
"mean_token_accuracy": 0.1936095252633095,
|
|
"num_tokens": 12655125.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.693441438674927,
|
|
"epoch": 5.984099699183498,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023210786690986646,
|
|
"loss": 5.2917,
|
|
"mean_token_accuracy": 0.19222740679979325,
|
|
"num_tokens": 12663465.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.652088069915772,
|
|
"epoch": 5.988397077782553,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00023178152829079712,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.20337795913219453,
|
|
"num_tokens": 12671573.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.599085283279419,
|
|
"epoch": 5.992694456381607,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00023145528404359562,
|
|
"loss": 5.1467,
|
|
"mean_token_accuracy": 0.21071504205465316,
|
|
"num_tokens": 12680865.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.7047466278076175,
|
|
"epoch": 5.996991834980662,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002311291348806492,
|
|
"loss": 5.2065,
|
|
"mean_token_accuracy": 0.19738373756408692,
|
|
"num_tokens": 12689790.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.610792954762776,
|
|
"epoch": 6.000859475719811,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002308030815141372,
|
|
"loss": 5.1243,
|
|
"mean_token_accuracy": 0.21220017638471392,
|
|
"num_tokens": 12697458.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.651318025588989,
|
|
"epoch": 6.005156854318866,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023047712465602976,
|
|
"loss": 5.0266,
|
|
"mean_token_accuracy": 0.2155851051211357,
|
|
"num_tokens": 12707122.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.638798952102661,
|
|
"epoch": 6.00945423291792,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023015126501808641,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.21842863410711288,
|
|
"num_tokens": 12715887.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.652589511871338,
|
|
"epoch": 6.013751611516975,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022982550331185437,
|
|
"loss": 4.9759,
|
|
"mean_token_accuracy": 0.22173826694488524,
|
|
"num_tokens": 12724630.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.013751611516975,
|
|
"eval_entropy": 5.561274541390909,
|
|
"eval_loss": 5.982907295227051,
|
|
"eval_mean_token_accuracy": 0.1741570174056399,
|
|
"eval_num_tokens": 12724630.0,
|
|
"eval_runtime": 2.6302,
|
|
"eval_samples_per_second": 1349.348,
|
|
"eval_steps_per_second": 168.811,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.671926021575928,
|
|
"epoch": 6.0180489901160295,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022949984024866704,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.20257444232702254,
|
|
"num_tokens": 12734480.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.6760564804077145,
|
|
"epoch": 6.022346368715084,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002291742765396424,
|
|
"loss": 4.9296,
|
|
"mean_token_accuracy": 0.21724649220705033,
|
|
"num_tokens": 12742755.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.683782482147217,
|
|
"epoch": 6.0266437473141385,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022884881289568133,
|
|
"loss": 5.1259,
|
|
"mean_token_accuracy": 0.2034509226679802,
|
|
"num_tokens": 12751721.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.666363525390625,
|
|
"epoch": 6.030941125913193,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002285234500274665,
|
|
"loss": 4.9437,
|
|
"mean_token_accuracy": 0.22417264729738234,
|
|
"num_tokens": 12759982.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.759365606307983,
|
|
"epoch": 6.035238504512248,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00022819818864546016,
|
|
"loss": 5.0693,
|
|
"mean_token_accuracy": 0.20940571576356887,
|
|
"num_tokens": 12768874.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.6370748519897464,
|
|
"epoch": 6.039535883111302,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022787302945990345,
|
|
"loss": 5.003,
|
|
"mean_token_accuracy": 0.217072494328022,
|
|
"num_tokens": 12777613.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.6079377174377445,
|
|
"epoch": 6.043833261710357,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00022754797318081383,
|
|
"loss": 5.0501,
|
|
"mean_token_accuracy": 0.21487515419721603,
|
|
"num_tokens": 12785436.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.691811275482178,
|
|
"epoch": 6.048130640309411,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00022722302051798442,
|
|
"loss": 5.145,
|
|
"mean_token_accuracy": 0.1987710788846016,
|
|
"num_tokens": 12795334.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.6396173477172855,
|
|
"epoch": 6.052428018908466,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002268981721809819,
|
|
"loss": 4.9885,
|
|
"mean_token_accuracy": 0.21851232498884202,
|
|
"num_tokens": 12804579.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.666747999191284,
|
|
"epoch": 6.05672539750752,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002265734288791451,
|
|
"loss": 5.1504,
|
|
"mean_token_accuracy": 0.19720997363328935,
|
|
"num_tokens": 12813557.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.665329170227051,
|
|
"epoch": 6.061022776106575,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022624879132158377,
|
|
"loss": 5.103,
|
|
"mean_token_accuracy": 0.20465276390314102,
|
|
"num_tokens": 12822588.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.720922613143921,
|
|
"epoch": 6.0653201547056295,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022592426021717654,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.2059031367301941,
|
|
"num_tokens": 12832543.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.734258794784546,
|
|
"epoch": 6.069617533304684,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002255998362745696,
|
|
"loss": 5.0745,
|
|
"mean_token_accuracy": 0.21017536669969558,
|
|
"num_tokens": 12841237.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.773473882675171,
|
|
"epoch": 6.073914911903739,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022527552020217513,
|
|
"loss": 5.188,
|
|
"mean_token_accuracy": 0.20581013411283494,
|
|
"num_tokens": 12850626.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.640860271453858,
|
|
"epoch": 6.078212290502793,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002249513127081697,
|
|
"loss": 5.1192,
|
|
"mean_token_accuracy": 0.20671833753585817,
|
|
"num_tokens": 12859290.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.675141048431397,
|
|
"epoch": 6.082509669101848,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00022462721450049316,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.20815706104040146,
|
|
"num_tokens": 12867828.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.61317663192749,
|
|
"epoch": 6.086807047700902,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002243032262868464,
|
|
"loss": 5.0274,
|
|
"mean_token_accuracy": 0.20926202237606048,
|
|
"num_tokens": 12876186.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.657180738449097,
|
|
"epoch": 6.091104426299957,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00022397934877469,
|
|
"loss": 5.0444,
|
|
"mean_token_accuracy": 0.2117959350347519,
|
|
"num_tokens": 12884657.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.63351263999939,
|
|
"epoch": 6.095401804899011,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002236555826712432,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.20999282896518706,
|
|
"num_tokens": 12893760.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.670510768890381,
|
|
"epoch": 6.099699183498066,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022333192868348152,
|
|
"loss": 5.1277,
|
|
"mean_token_accuracy": 0.20132159292697907,
|
|
"num_tokens": 12903263.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.687745475769043,
|
|
"epoch": 6.10399656209712,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022300838751813606,
|
|
"loss": 5.0195,
|
|
"mean_token_accuracy": 0.21672120988368987,
|
|
"num_tokens": 12912497.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.710108852386474,
|
|
"epoch": 6.108293940696175,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022268495988169145,
|
|
"loss": 5.0791,
|
|
"mean_token_accuracy": 0.20718155801296234,
|
|
"num_tokens": 12921977.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.597428607940674,
|
|
"epoch": 6.1125913192952295,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00022236164648038433,
|
|
"loss": 5.0557,
|
|
"mean_token_accuracy": 0.20747778415679932,
|
|
"num_tokens": 12931845.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.7094532489776615,
|
|
"epoch": 6.116888697894285,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002220384480202019,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.2077478289604187,
|
|
"num_tokens": 12941613.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.751816511154175,
|
|
"epoch": 6.1211860764933395,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022171536520688046,
|
|
"loss": 5.0527,
|
|
"mean_token_accuracy": 0.2112714469432831,
|
|
"num_tokens": 12950570.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.693778657913208,
|
|
"epoch": 6.125483455092394,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00022139239874590362,
|
|
"loss": 5.0548,
|
|
"mean_token_accuracy": 0.20674584358930587,
|
|
"num_tokens": 12959576.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.625174570083618,
|
|
"epoch": 6.1297808336914485,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002210695493425013,
|
|
"loss": 5.0083,
|
|
"mean_token_accuracy": 0.2162788763642311,
|
|
"num_tokens": 12968333.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.6447060108184814,
|
|
"epoch": 6.134078212290503,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022074681770164735,
|
|
"loss": 5.0613,
|
|
"mean_token_accuracy": 0.20821176171302797,
|
|
"num_tokens": 12977213.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.521041965484619,
|
|
"epoch": 6.138375590889558,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00022042420452805868,
|
|
"loss": 4.9045,
|
|
"mean_token_accuracy": 0.23039672523736954,
|
|
"num_tokens": 12986289.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.676589679718018,
|
|
"epoch": 6.142672969488612,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00022010171052619365,
|
|
"loss": 4.9985,
|
|
"mean_token_accuracy": 0.21907657980918885,
|
|
"num_tokens": 12994532.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.663532781600952,
|
|
"epoch": 6.146970348087667,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00021977933640025,
|
|
"loss": 5.0705,
|
|
"mean_token_accuracy": 0.21151696443557738,
|
|
"num_tokens": 13003951.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.604747867584228,
|
|
"epoch": 6.151267726686721,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021945708285416434,
|
|
"loss": 5.0663,
|
|
"mean_token_accuracy": 0.2151520371437073,
|
|
"num_tokens": 13013123.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.658300876617432,
|
|
"epoch": 6.155565105285776,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002191349505916093,
|
|
"loss": 5.0963,
|
|
"mean_token_accuracy": 0.20420115888118745,
|
|
"num_tokens": 13021451.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.6494592189788815,
|
|
"epoch": 6.15986248388483,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021881294031599318,
|
|
"loss": 5.0039,
|
|
"mean_token_accuracy": 0.21728747487068176,
|
|
"num_tokens": 13030894.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.676177549362182,
|
|
"epoch": 6.164159862483885,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002184910527304576,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.21043265610933304,
|
|
"num_tokens": 13039930.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.698052024841308,
|
|
"epoch": 6.1684572410829395,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00021816928853787636,
|
|
"loss": 5.1259,
|
|
"mean_token_accuracy": 0.2038058191537857,
|
|
"num_tokens": 13048575.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.749299478530884,
|
|
"epoch": 6.172754619681994,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00021784764844085398,
|
|
"loss": 5.2018,
|
|
"mean_token_accuracy": 0.19360179007053374,
|
|
"num_tokens": 13057624.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.622776651382447,
|
|
"epoch": 6.177051998281049,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002175261331417238,
|
|
"loss": 4.995,
|
|
"mean_token_accuracy": 0.2219090238213539,
|
|
"num_tokens": 13066556.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.6489564895629885,
|
|
"epoch": 6.181349376880103,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021720474334254675,
|
|
"loss": 5.0243,
|
|
"mean_token_accuracy": 0.2155717834830284,
|
|
"num_tokens": 13075617.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.679793500900269,
|
|
"epoch": 6.185646755479158,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00021688347974510962,
|
|
"loss": 5.0573,
|
|
"mean_token_accuracy": 0.21591018587350846,
|
|
"num_tokens": 13084528.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.559691286087036,
|
|
"epoch": 6.189944134078212,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00021656234305092377,
|
|
"loss": 5.0093,
|
|
"mean_token_accuracy": 0.2171579360961914,
|
|
"num_tokens": 13093841.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.59034571647644,
|
|
"epoch": 6.194241512677267,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002162413339612234,
|
|
"loss": 5.1035,
|
|
"mean_token_accuracy": 0.2077306792140007,
|
|
"num_tokens": 13103104.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.704719686508179,
|
|
"epoch": 6.198538891276321,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021592045317696406,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.20979773104190827,
|
|
"num_tokens": 13112721.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.7365556240081785,
|
|
"epoch": 6.202836269875376,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00021559970139882102,
|
|
"loss": 5.1531,
|
|
"mean_token_accuracy": 0.20669684559106827,
|
|
"num_tokens": 13121949.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.695716619491577,
|
|
"epoch": 6.20713364847443,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002152790793271881,
|
|
"loss": 5.1377,
|
|
"mean_token_accuracy": 0.20416705161333085,
|
|
"num_tokens": 13130693.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.733968305587768,
|
|
"epoch": 6.211431027073485,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021495858766217558,
|
|
"loss": 5.1416,
|
|
"mean_token_accuracy": 0.2000860333442688,
|
|
"num_tokens": 13140438.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.662462329864502,
|
|
"epoch": 6.2157284056725395,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021463822710360932,
|
|
"loss": 5.0405,
|
|
"mean_token_accuracy": 0.20904087126255036,
|
|
"num_tokens": 13149621.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.633723020553589,
|
|
"epoch": 6.220025784271594,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00021431799835102867,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.21213067471981048,
|
|
"num_tokens": 13158366.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.702373313903808,
|
|
"epoch": 6.224323162870649,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021399790210368524,
|
|
"loss": 5.1539,
|
|
"mean_token_accuracy": 0.2022456258535385,
|
|
"num_tokens": 13167303.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.680520009994507,
|
|
"epoch": 6.228620541469703,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00021367793906054133,
|
|
"loss": 5.0988,
|
|
"mean_token_accuracy": 0.21145589500665665,
|
|
"num_tokens": 13175276.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.727468729019165,
|
|
"epoch": 6.232917920068758,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00021335810992026823,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.20255154073238374,
|
|
"num_tokens": 13184352.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.64649806022644,
|
|
"epoch": 6.237215298667812,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00021303841538124497,
|
|
"loss": 5.0406,
|
|
"mean_token_accuracy": 0.21567850410938263,
|
|
"num_tokens": 13192772.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.671575975418091,
|
|
"epoch": 6.241512677266867,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021271885614155685,
|
|
"loss": 5.1011,
|
|
"mean_token_accuracy": 0.21192974895238875,
|
|
"num_tokens": 13201574.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.647490215301514,
|
|
"epoch": 6.245810055865922,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002123994328989932,
|
|
"loss": 5.0744,
|
|
"mean_token_accuracy": 0.2082773983478546,
|
|
"num_tokens": 13211572.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.682449388504028,
|
|
"epoch": 6.250107434464977,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00021208014635104688,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.19993288516998292,
|
|
"num_tokens": 13220614.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.671002006530761,
|
|
"epoch": 6.254404813064031,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00021176099719491209,
|
|
"loss": 5.0697,
|
|
"mean_token_accuracy": 0.21990010589361192,
|
|
"num_tokens": 13229484.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.647109079360962,
|
|
"epoch": 6.258702191663086,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00021144198612748312,
|
|
"loss": 5.1005,
|
|
"mean_token_accuracy": 0.2097909465432167,
|
|
"num_tokens": 13239649.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.675845670700073,
|
|
"epoch": 6.26299957026214,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00021112311384535243,
|
|
"loss": 5.002,
|
|
"mean_token_accuracy": 0.21777600198984146,
|
|
"num_tokens": 13248199.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.698729467391968,
|
|
"epoch": 6.267296948861195,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021080438104480976,
|
|
"loss": 5.1105,
|
|
"mean_token_accuracy": 0.20624627619981767,
|
|
"num_tokens": 13257254.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.6558455467224125,
|
|
"epoch": 6.2715943274602495,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021048578842184019,
|
|
"loss": 5.0926,
|
|
"mean_token_accuracy": 0.20991834551095961,
|
|
"num_tokens": 13266272.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.621621942520141,
|
|
"epoch": 6.275891706059304,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00021016733667212245,
|
|
"loss": 5.067,
|
|
"mean_token_accuracy": 0.2159287989139557,
|
|
"num_tokens": 13274871.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.722089529037476,
|
|
"epoch": 6.280189084658359,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020984902649102806,
|
|
"loss": 5.1561,
|
|
"mean_token_accuracy": 0.20660688877105712,
|
|
"num_tokens": 13284844.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.713557720184326,
|
|
"epoch": 6.284486463257413,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00020953085857361924,
|
|
"loss": 5.1073,
|
|
"mean_token_accuracy": 0.20475954711437225,
|
|
"num_tokens": 13293977.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.705980253219605,
|
|
"epoch": 6.288783841856468,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00020921283361464754,
|
|
"loss": 5.1604,
|
|
"mean_token_accuracy": 0.2033306688070297,
|
|
"num_tokens": 13303359.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.729022789001465,
|
|
"epoch": 6.293081220455522,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020889495230855232,
|
|
"loss": 5.1275,
|
|
"mean_token_accuracy": 0.21200229972600937,
|
|
"num_tokens": 13311965.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.727298927307129,
|
|
"epoch": 6.297378599054577,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020857721534945923,
|
|
"loss": 5.1635,
|
|
"mean_token_accuracy": 0.20653231739997863,
|
|
"num_tokens": 13321784.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.673910522460938,
|
|
"epoch": 6.301675977653631,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002082596234311789,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.2086361140012741,
|
|
"num_tokens": 13330141.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.576520490646362,
|
|
"epoch": 6.305973356252686,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002079421772472051,
|
|
"loss": 5.076,
|
|
"mean_token_accuracy": 0.20600971281528474,
|
|
"num_tokens": 13339303.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.68666639328003,
|
|
"epoch": 6.31027073485174,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002076248774907134,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.21266597807407378,
|
|
"num_tokens": 13348574.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.7680335521698,
|
|
"epoch": 6.314568113450795,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00020730772485455962,
|
|
"loss": 5.1025,
|
|
"mean_token_accuracy": 0.21058187633752823,
|
|
"num_tokens": 13356770.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.712491703033447,
|
|
"epoch": 6.3188654920498495,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002069907200312785,
|
|
"loss": 5.1469,
|
|
"mean_token_accuracy": 0.20520444661378862,
|
|
"num_tokens": 13366973.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.688606786727905,
|
|
"epoch": 6.323162870648904,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020667386371308162,
|
|
"loss": 5.1479,
|
|
"mean_token_accuracy": 0.20392710566520691,
|
|
"num_tokens": 13375849.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.600723791122436,
|
|
"epoch": 6.327460249247959,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020635715659185673,
|
|
"loss": 5.0275,
|
|
"mean_token_accuracy": 0.2151519849896431,
|
|
"num_tokens": 13384051.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.685665464401245,
|
|
"epoch": 6.331757627847013,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020604059935916551,
|
|
"loss": 5.1584,
|
|
"mean_token_accuracy": 0.20562170892953874,
|
|
"num_tokens": 13393194.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.6784992694854735,
|
|
"epoch": 6.336055006446068,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020572419270624255,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.2097678080201149,
|
|
"num_tokens": 13402701.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.602465009689331,
|
|
"epoch": 6.340352385045122,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020540793732399339,
|
|
"loss": 4.9696,
|
|
"mean_token_accuracy": 0.21893648952245712,
|
|
"num_tokens": 13411583.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.653471374511719,
|
|
"epoch": 6.344649763644177,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020509183390299325,
|
|
"loss": 5.0942,
|
|
"mean_token_accuracy": 0.20453108251094818,
|
|
"num_tokens": 13420787.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.657783269882202,
|
|
"epoch": 6.348947142243231,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020477588313348594,
|
|
"loss": 5.1254,
|
|
"mean_token_accuracy": 0.20339320451021195,
|
|
"num_tokens": 13428903.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.66182770729065,
|
|
"epoch": 6.353244520842286,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020446008570538154,
|
|
"loss": 5.0998,
|
|
"mean_token_accuracy": 0.20520801991224288,
|
|
"num_tokens": 13438444.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.621269607543946,
|
|
"epoch": 6.35754189944134,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002041444423082554,
|
|
"loss": 5.1113,
|
|
"mean_token_accuracy": 0.20502590984106064,
|
|
"num_tokens": 13447188.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.696452569961548,
|
|
"epoch": 6.361839278040395,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00020382895363134652,
|
|
"loss": 5.105,
|
|
"mean_token_accuracy": 0.20173575729131699,
|
|
"num_tokens": 13455980.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.725162410736084,
|
|
"epoch": 6.3661366566394495,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020351362036355602,
|
|
"loss": 5.0827,
|
|
"mean_token_accuracy": 0.21074196845293044,
|
|
"num_tokens": 13465029.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.727134752273559,
|
|
"epoch": 6.370434035238505,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002031984431934459,
|
|
"loss": 5.1413,
|
|
"mean_token_accuracy": 0.2065821960568428,
|
|
"num_tokens": 13474366.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.718928623199463,
|
|
"epoch": 6.3747314138375595,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00020288342280923695,
|
|
"loss": 5.1125,
|
|
"mean_token_accuracy": 0.20159267485141755,
|
|
"num_tokens": 13483186.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.7256622314453125,
|
|
"epoch": 6.379028792436614,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020256855989880785,
|
|
"loss": 5.1839,
|
|
"mean_token_accuracy": 0.19618037790060044,
|
|
"num_tokens": 13492066.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.660233354568481,
|
|
"epoch": 6.383326171035669,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00020225385514969336,
|
|
"loss": 5.0842,
|
|
"mean_token_accuracy": 0.2043602004647255,
|
|
"num_tokens": 13501671.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.644028997421264,
|
|
"epoch": 6.387623549634723,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00020193930924908277,
|
|
"loss": 5.0357,
|
|
"mean_token_accuracy": 0.21109388917684554,
|
|
"num_tokens": 13510323.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.678756046295166,
|
|
"epoch": 6.391920928233778,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00020162492288381867,
|
|
"loss": 5.0686,
|
|
"mean_token_accuracy": 0.2101046025753021,
|
|
"num_tokens": 13518658.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.666945838928223,
|
|
"epoch": 6.396218306832832,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002013106967403953,
|
|
"loss": 5.1445,
|
|
"mean_token_accuracy": 0.20326807647943496,
|
|
"num_tokens": 13528053.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.706927251815796,
|
|
"epoch": 6.400515685431887,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002009966315049569,
|
|
"loss": 5.0617,
|
|
"mean_token_accuracy": 0.21021295487880706,
|
|
"num_tokens": 13537125.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.704835653305054,
|
|
"epoch": 6.404813064030941,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002006827278632964,
|
|
"loss": 5.1734,
|
|
"mean_token_accuracy": 0.20163009017705918,
|
|
"num_tokens": 13547859.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.686561059951782,
|
|
"epoch": 6.409110442629996,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020036898650085377,
|
|
"loss": 5.1396,
|
|
"mean_token_accuracy": 0.2101307302713394,
|
|
"num_tokens": 13557545.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.715054607391357,
|
|
"epoch": 6.41340782122905,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00020005540810271493,
|
|
"loss": 5.1673,
|
|
"mean_token_accuracy": 0.1998737022280693,
|
|
"num_tokens": 13566281.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.6541985988616945,
|
|
"epoch": 6.417705199828105,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00019974199335360976,
|
|
"loss": 5.0299,
|
|
"mean_token_accuracy": 0.21672433465719224,
|
|
"num_tokens": 13575461.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.650461292266845,
|
|
"epoch": 6.4220025784271595,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019942874293791068,
|
|
"loss": 5.0466,
|
|
"mean_token_accuracy": 0.2165360778570175,
|
|
"num_tokens": 13585107.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.700667476654052,
|
|
"epoch": 6.426299957026214,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019911565753963145,
|
|
"loss": 5.0504,
|
|
"mean_token_accuracy": 0.21274675875902177,
|
|
"num_tokens": 13593787.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.692003154754639,
|
|
"epoch": 6.430597335625269,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001988027378424254,
|
|
"loss": 5.1205,
|
|
"mean_token_accuracy": 0.20692434906959534,
|
|
"num_tokens": 13602956.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.647941064834595,
|
|
"epoch": 6.434894714224323,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019848998452958429,
|
|
"loss": 5.0945,
|
|
"mean_token_accuracy": 0.20780716687440873,
|
|
"num_tokens": 13612380.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.603743219375611,
|
|
"epoch": 6.439192092823378,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019817739828403602,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.20554549098014832,
|
|
"num_tokens": 13622049.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.697331953048706,
|
|
"epoch": 6.443489471422432,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019786497978834422,
|
|
"loss": 5.1271,
|
|
"mean_token_accuracy": 0.19796979129314424,
|
|
"num_tokens": 13631497.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.443489471422432,
|
|
"eval_entropy": 5.555225222497373,
|
|
"eval_loss": 5.974055767059326,
|
|
"eval_mean_token_accuracy": 0.17487258410400097,
|
|
"eval_num_tokens": 13631497.0,
|
|
"eval_runtime": 2.6424,
|
|
"eval_samples_per_second": 1343.097,
|
|
"eval_steps_per_second": 168.029,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.687777042388916,
|
|
"epoch": 6.447786850021487,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.00019755272972470602,
|
|
"loss": 5.0961,
|
|
"mean_token_accuracy": 0.19693411886692047,
|
|
"num_tokens": 13639767.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.684772157669068,
|
|
"epoch": 6.452084228620541,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00019724064877495057,
|
|
"loss": 5.0978,
|
|
"mean_token_accuracy": 0.200966976583004,
|
|
"num_tokens": 13649255.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.746344041824341,
|
|
"epoch": 6.456381607219596,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00019692873762053808,
|
|
"loss": 5.1778,
|
|
"mean_token_accuracy": 0.20268609523773193,
|
|
"num_tokens": 13657764.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.684163284301758,
|
|
"epoch": 6.46067898581865,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019661699694255785,
|
|
"loss": 5.1358,
|
|
"mean_token_accuracy": 0.20156339257955552,
|
|
"num_tokens": 13667922.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.695675611495972,
|
|
"epoch": 6.464976364417705,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00019630542742172692,
|
|
"loss": 5.1592,
|
|
"mean_token_accuracy": 0.203003191947937,
|
|
"num_tokens": 13678196.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.680872869491577,
|
|
"epoch": 6.4692737430167595,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00019599402973838854,
|
|
"loss": 5.1825,
|
|
"mean_token_accuracy": 0.2007237359881401,
|
|
"num_tokens": 13687605.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.642401456832886,
|
|
"epoch": 6.473571121615814,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001956828045725107,
|
|
"loss": 5.0418,
|
|
"mean_token_accuracy": 0.21417637914419174,
|
|
"num_tokens": 13695790.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.66036901473999,
|
|
"epoch": 6.477868500214869,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001953717526036849,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.20346380025148392,
|
|
"num_tokens": 13705608.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.6280584812164305,
|
|
"epoch": 6.482165878813923,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00019506087451112437,
|
|
"loss": 5.0255,
|
|
"mean_token_accuracy": 0.21617298871278762,
|
|
"num_tokens": 13714286.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.687048149108887,
|
|
"epoch": 6.486463257412978,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00019475017097366244,
|
|
"loss": 5.0498,
|
|
"mean_token_accuracy": 0.2129441723227501,
|
|
"num_tokens": 13722575.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.685393238067627,
|
|
"epoch": 6.490760636012032,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00019443964266975156,
|
|
"loss": 5.1092,
|
|
"mean_token_accuracy": 0.2061144933104515,
|
|
"num_tokens": 13732342.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.615344905853272,
|
|
"epoch": 6.495058014611088,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001941292902774614,
|
|
"loss": 5.0115,
|
|
"mean_token_accuracy": 0.21612818092107772,
|
|
"num_tokens": 13741386.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.657469940185547,
|
|
"epoch": 6.499355393210142,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019381911447447742,
|
|
"loss": 5.0994,
|
|
"mean_token_accuracy": 0.21311458349227905,
|
|
"num_tokens": 13750755.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.675438404083252,
|
|
"epoch": 6.503652771809197,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00019350911593809977,
|
|
"loss": 5.0267,
|
|
"mean_token_accuracy": 0.21482536643743516,
|
|
"num_tokens": 13759903.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.631055879592895,
|
|
"epoch": 6.507950150408251,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019319929534524128,
|
|
"loss": 5.0188,
|
|
"mean_token_accuracy": 0.21048175990581514,
|
|
"num_tokens": 13768066.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.617418003082276,
|
|
"epoch": 6.512247529007306,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00019288965337242636,
|
|
"loss": 5.0696,
|
|
"mean_token_accuracy": 0.21266268491744994,
|
|
"num_tokens": 13778010.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.6921099662780765,
|
|
"epoch": 6.51654490760636,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00019258019069578924,
|
|
"loss": 5.1504,
|
|
"mean_token_accuracy": 0.2000333473086357,
|
|
"num_tokens": 13787415.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.646253776550293,
|
|
"epoch": 6.520842286205415,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019227090799107266,
|
|
"loss": 5.0497,
|
|
"mean_token_accuracy": 0.21170607954263687,
|
|
"num_tokens": 13796937.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.759700584411621,
|
|
"epoch": 6.5251396648044695,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001919618059336265,
|
|
"loss": 5.1183,
|
|
"mean_token_accuracy": 0.2070131078362465,
|
|
"num_tokens": 13805759.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.685678386688233,
|
|
"epoch": 6.529437043403524,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019165288519840617,
|
|
"loss": 5.0737,
|
|
"mean_token_accuracy": 0.21074699163436889,
|
|
"num_tokens": 13814915.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.622214126586914,
|
|
"epoch": 6.533734422002579,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000191344146459971,
|
|
"loss": 5.0842,
|
|
"mean_token_accuracy": 0.20493489056825637,
|
|
"num_tokens": 13824334.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.661040496826172,
|
|
"epoch": 6.538031800601633,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00019103559039248302,
|
|
"loss": 5.0845,
|
|
"mean_token_accuracy": 0.20999274551868438,
|
|
"num_tokens": 13834124.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.615899658203125,
|
|
"epoch": 6.542329179200688,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0001907272176697052,
|
|
"loss": 5.0044,
|
|
"mean_token_accuracy": 0.22054746001958847,
|
|
"num_tokens": 13843240.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.724200677871704,
|
|
"epoch": 6.546626557799742,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019041902896500059,
|
|
"loss": 5.1669,
|
|
"mean_token_accuracy": 0.20176565796136856,
|
|
"num_tokens": 13852837.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.730016803741455,
|
|
"epoch": 6.550923936398797,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019011102495132993,
|
|
"loss": 5.1086,
|
|
"mean_token_accuracy": 0.2046348661184311,
|
|
"num_tokens": 13861650.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.66640944480896,
|
|
"epoch": 6.555221314997851,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018980320630125104,
|
|
"loss": 5.1326,
|
|
"mean_token_accuracy": 0.20683347433805466,
|
|
"num_tokens": 13871939.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.798937892913818,
|
|
"epoch": 6.559518693596906,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00018949557368691666,
|
|
"loss": 5.1926,
|
|
"mean_token_accuracy": 0.1953928977251053,
|
|
"num_tokens": 13881132.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.635640430450439,
|
|
"epoch": 6.56381607219596,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00018918812778007343,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.21129678189754486,
|
|
"num_tokens": 13890476.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.600541019439698,
|
|
"epoch": 6.568113450795015,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018888086925206054,
|
|
"loss": 5.0575,
|
|
"mean_token_accuracy": 0.20954867899417878,
|
|
"num_tokens": 13900720.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.636783790588379,
|
|
"epoch": 6.5724108293940695,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00018857379877380763,
|
|
"loss": 5.1205,
|
|
"mean_token_accuracy": 0.20689990371465683,
|
|
"num_tokens": 13909461.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.694855737686157,
|
|
"epoch": 6.576708207993124,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018826691701583404,
|
|
"loss": 5.0753,
|
|
"mean_token_accuracy": 0.21015714406967162,
|
|
"num_tokens": 13918136.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.6900440692901615,
|
|
"epoch": 6.581005586592179,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018796022464824663,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.19822703748941423,
|
|
"num_tokens": 13926965.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.741218090057373,
|
|
"epoch": 6.585302965191233,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018765372234073912,
|
|
"loss": 5.2145,
|
|
"mean_token_accuracy": 0.1936648204922676,
|
|
"num_tokens": 13936468.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.673659038543701,
|
|
"epoch": 6.589600343790288,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00018734741076259005,
|
|
"loss": 5.0394,
|
|
"mean_token_accuracy": 0.21618155092000962,
|
|
"num_tokens": 13945424.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.678310775756836,
|
|
"epoch": 6.593897722389342,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018704129058266152,
|
|
"loss": 5.0889,
|
|
"mean_token_accuracy": 0.2080785810947418,
|
|
"num_tokens": 13955129.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.731112909317017,
|
|
"epoch": 6.598195100988397,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018673536246939743,
|
|
"loss": 5.1767,
|
|
"mean_token_accuracy": 0.19521082937717438,
|
|
"num_tokens": 13964316.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.6261883735656735,
|
|
"epoch": 6.602492479587451,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018642962709082274,
|
|
"loss": 5.0766,
|
|
"mean_token_accuracy": 0.21126716881990432,
|
|
"num_tokens": 13973173.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.6725287437438965,
|
|
"epoch": 6.606789858186506,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018612408511454103,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.21258099526166915,
|
|
"num_tokens": 13981531.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.678965473175049,
|
|
"epoch": 6.61108723678556,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00018581873720773423,
|
|
"loss": 5.1419,
|
|
"mean_token_accuracy": 0.20654426217079164,
|
|
"num_tokens": 13990202.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.723487854003906,
|
|
"epoch": 6.615384615384615,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018551358403715989,
|
|
"loss": 5.0771,
|
|
"mean_token_accuracy": 0.21180738210678102,
|
|
"num_tokens": 13998851.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.6786407947540285,
|
|
"epoch": 6.61968199398367,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018520862626915052,
|
|
"loss": 5.0794,
|
|
"mean_token_accuracy": 0.20810818076133727,
|
|
"num_tokens": 14008120.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.661171817779541,
|
|
"epoch": 6.623979372582724,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018490386456961223,
|
|
"loss": 5.1632,
|
|
"mean_token_accuracy": 0.2010783076286316,
|
|
"num_tokens": 14017435.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.674332094192505,
|
|
"epoch": 6.6282767511817795,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001845992996040224,
|
|
"loss": 5.1846,
|
|
"mean_token_accuracy": 0.19881236404180527,
|
|
"num_tokens": 14026260.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.786124563217163,
|
|
"epoch": 6.632574129780834,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018429493203742946,
|
|
"loss": 5.169,
|
|
"mean_token_accuracy": 0.2001948520541191,
|
|
"num_tokens": 14034854.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.666089010238648,
|
|
"epoch": 6.636871508379889,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018399076253445052,
|
|
"loss": 5.0943,
|
|
"mean_token_accuracy": 0.20444100797176362,
|
|
"num_tokens": 14043486.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.645094919204712,
|
|
"epoch": 6.641168886978943,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018368679175927012,
|
|
"loss": 5.1659,
|
|
"mean_token_accuracy": 0.20071545094251633,
|
|
"num_tokens": 14052903.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.633981800079345,
|
|
"epoch": 6.645466265577998,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00018338302037563885,
|
|
"loss": 5.0828,
|
|
"mean_token_accuracy": 0.2124032959342003,
|
|
"num_tokens": 14061679.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.687230825424194,
|
|
"epoch": 6.649763644177052,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018307944904687211,
|
|
"loss": 5.1393,
|
|
"mean_token_accuracy": 0.2071155697107315,
|
|
"num_tokens": 14070549.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.623162698745728,
|
|
"epoch": 6.654061022776107,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001827760784358483,
|
|
"loss": 5.0046,
|
|
"mean_token_accuracy": 0.21896424293518066,
|
|
"num_tokens": 14079321.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.683493566513062,
|
|
"epoch": 6.658358401375161,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018247290920500776,
|
|
"loss": 5.131,
|
|
"mean_token_accuracy": 0.2007245570421219,
|
|
"num_tokens": 14087951.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.5430505752563475,
|
|
"epoch": 6.662655779974216,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018216994201635062,
|
|
"loss": 5.0439,
|
|
"mean_token_accuracy": 0.21624985188245774,
|
|
"num_tokens": 14097563.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.707266330718994,
|
|
"epoch": 6.66695315857327,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018186717753143633,
|
|
"loss": 5.1309,
|
|
"mean_token_accuracy": 0.20179227590560914,
|
|
"num_tokens": 14106961.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.693455553054809,
|
|
"epoch": 6.671250537172325,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018156461641138133,
|
|
"loss": 5.1332,
|
|
"mean_token_accuracy": 0.20799991339445115,
|
|
"num_tokens": 14115854.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.731644248962402,
|
|
"epoch": 6.6755479157713795,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018126225931685836,
|
|
"loss": 5.1542,
|
|
"mean_token_accuracy": 0.20515426248311996,
|
|
"num_tokens": 14124778.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.659886074066162,
|
|
"epoch": 6.679845294370434,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018096010690809444,
|
|
"loss": 5.0599,
|
|
"mean_token_accuracy": 0.2189943462610245,
|
|
"num_tokens": 14133708.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.5866025447845455,
|
|
"epoch": 6.684142672969489,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018065815984486962,
|
|
"loss": 4.9916,
|
|
"mean_token_accuracy": 0.221568363904953,
|
|
"num_tokens": 14142885.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.637514257431031,
|
|
"epoch": 6.688440051568543,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018035641878651548,
|
|
"loss": 5.1077,
|
|
"mean_token_accuracy": 0.20211369246244432,
|
|
"num_tokens": 14151156.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.694208002090454,
|
|
"epoch": 6.692737430167598,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00018005488439191408,
|
|
"loss": 5.1102,
|
|
"mean_token_accuracy": 0.20702400356531142,
|
|
"num_tokens": 14160482.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.70957579612732,
|
|
"epoch": 6.697034808766652,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001797535573194959,
|
|
"loss": 5.0699,
|
|
"mean_token_accuracy": 0.2134239360690117,
|
|
"num_tokens": 14169637.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.619124984741211,
|
|
"epoch": 6.701332187365707,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001794524382272389,
|
|
"loss": 5.1359,
|
|
"mean_token_accuracy": 0.19880518764257432,
|
|
"num_tokens": 14179215.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.722444200515747,
|
|
"epoch": 6.705629565964761,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001791515277726667,
|
|
"loss": 5.186,
|
|
"mean_token_accuracy": 0.204307459294796,
|
|
"num_tokens": 14188484.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.63198709487915,
|
|
"epoch": 6.709926944563816,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017885082661284763,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.2148973301053047,
|
|
"num_tokens": 14197275.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.664177322387696,
|
|
"epoch": 6.71422432316287,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017855033540439274,
|
|
"loss": 5.0674,
|
|
"mean_token_accuracy": 0.21290301084518432,
|
|
"num_tokens": 14206260.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.699127054214477,
|
|
"epoch": 6.718521701761925,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00017825005480345463,
|
|
"loss": 5.1645,
|
|
"mean_token_accuracy": 0.20044873803853988,
|
|
"num_tokens": 14215758.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.6080128192901615,
|
|
"epoch": 6.7228190803609795,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017794998546572627,
|
|
"loss": 5.0721,
|
|
"mean_token_accuracy": 0.21099960803985596,
|
|
"num_tokens": 14225147.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.654290437698364,
|
|
"epoch": 6.727116458960034,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001776501280464391,
|
|
"loss": 5.1085,
|
|
"mean_token_accuracy": 0.20604175925254822,
|
|
"num_tokens": 14233567.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.638417053222656,
|
|
"epoch": 6.731413837559089,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017735048320036197,
|
|
"loss": 5.013,
|
|
"mean_token_accuracy": 0.21838999390602112,
|
|
"num_tokens": 14242101.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.66321325302124,
|
|
"epoch": 6.735711216158143,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017705105158179917,
|
|
"loss": 5.0593,
|
|
"mean_token_accuracy": 0.20810490399599074,
|
|
"num_tokens": 14251502.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.6178391456604,
|
|
"epoch": 6.740008594757198,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017675183384458987,
|
|
"loss": 5.0047,
|
|
"mean_token_accuracy": 0.21956098824739456,
|
|
"num_tokens": 14260059.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.740962076187134,
|
|
"epoch": 6.744305973356253,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017645283064210616,
|
|
"loss": 5.0772,
|
|
"mean_token_accuracy": 0.20966309756040574,
|
|
"num_tokens": 14268672.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.60942063331604,
|
|
"epoch": 6.748603351955307,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00017615404262725132,
|
|
"loss": 4.9487,
|
|
"mean_token_accuracy": 0.2202254354953766,
|
|
"num_tokens": 14277636.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.574077415466308,
|
|
"epoch": 6.752900730554362,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017585547045245885,
|
|
"loss": 5.0043,
|
|
"mean_token_accuracy": 0.2144310638308525,
|
|
"num_tokens": 14286417.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.6434245109558105,
|
|
"epoch": 6.757198109153417,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00017555711476969138,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.21289074718952178,
|
|
"num_tokens": 14294725.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.623466348648071,
|
|
"epoch": 6.761495487752471,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00017525897623043806,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.21595387309789657,
|
|
"num_tokens": 14303228.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.677674961090088,
|
|
"epoch": 6.765792866351526,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00017496105548571472,
|
|
"loss": 5.0948,
|
|
"mean_token_accuracy": 0.2045324847102165,
|
|
"num_tokens": 14312334.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.683128929138183,
|
|
"epoch": 6.77009024495058,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017466335318606086,
|
|
"loss": 5.1423,
|
|
"mean_token_accuracy": 0.2102939024567604,
|
|
"num_tokens": 14321635.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.754555797576904,
|
|
"epoch": 6.774387623549635,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017436586998153947,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.2017458513379097,
|
|
"num_tokens": 14332456.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.681694078445434,
|
|
"epoch": 6.7786850021486895,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017406860652173495,
|
|
"loss": 5.0461,
|
|
"mean_token_accuracy": 0.21011338680982589,
|
|
"num_tokens": 14341594.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.695341873168945,
|
|
"epoch": 6.782982380747744,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017377156345575176,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.20596107095479965,
|
|
"num_tokens": 14351123.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.6329017162323,
|
|
"epoch": 6.787279759346799,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017347474143221338,
|
|
"loss": 5.0704,
|
|
"mean_token_accuracy": 0.21130015105009078,
|
|
"num_tokens": 14360586.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.645985794067383,
|
|
"epoch": 6.791577137945853,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017317814109926044,
|
|
"loss": 5.0903,
|
|
"mean_token_accuracy": 0.2090660884976387,
|
|
"num_tokens": 14370101.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.698221969604492,
|
|
"epoch": 6.795874516544908,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001728817631045495,
|
|
"loss": 5.1689,
|
|
"mean_token_accuracy": 0.20833769887685777,
|
|
"num_tokens": 14380255.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.659369039535522,
|
|
"epoch": 6.800171895143962,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001725856080952516,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.20738712698221207,
|
|
"num_tokens": 14389563.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.605033588409424,
|
|
"epoch": 6.804469273743017,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001722896767180509,
|
|
"loss": 5.0057,
|
|
"mean_token_accuracy": 0.21646001040935517,
|
|
"num_tokens": 14398582.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.575069332122803,
|
|
"epoch": 6.808766652342071,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017199396961914334,
|
|
"loss": 5.1066,
|
|
"mean_token_accuracy": 0.2053375482559204,
|
|
"num_tokens": 14407739.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.727594900131225,
|
|
"epoch": 6.813064030941126,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00017169848744423506,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.20556814223527908,
|
|
"num_tokens": 14417624.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.736045265197754,
|
|
"epoch": 6.81736140954018,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017140323083854076,
|
|
"loss": 5.1717,
|
|
"mean_token_accuracy": 0.1974081352353096,
|
|
"num_tokens": 14427446.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.682717037200928,
|
|
"epoch": 6.821658788139235,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017110820044678317,
|
|
"loss": 5.0629,
|
|
"mean_token_accuracy": 0.2172720566391945,
|
|
"num_tokens": 14437274.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.616380405426026,
|
|
"epoch": 6.8259561667382895,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017081339691319054,
|
|
"loss": 5.0184,
|
|
"mean_token_accuracy": 0.20874297469854355,
|
|
"num_tokens": 14445786.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.629858827590942,
|
|
"epoch": 6.830253545337344,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017051882088149612,
|
|
"loss": 5.0552,
|
|
"mean_token_accuracy": 0.21024104207754135,
|
|
"num_tokens": 14454571.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.651091384887695,
|
|
"epoch": 6.834550923936399,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017022447299493599,
|
|
"loss": 5.0813,
|
|
"mean_token_accuracy": 0.20949982851743698,
|
|
"num_tokens": 14463645.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.6449785232543945,
|
|
"epoch": 6.838848302535453,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016993035389624854,
|
|
"loss": 5.0135,
|
|
"mean_token_accuracy": 0.2239181786775589,
|
|
"num_tokens": 14472737.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.697493553161621,
|
|
"epoch": 6.843145681134508,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001696364642276722,
|
|
"loss": 5.1244,
|
|
"mean_token_accuracy": 0.19821999371051788,
|
|
"num_tokens": 14482053.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.71565957069397,
|
|
"epoch": 6.847443059733562,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00016934280463094448,
|
|
"loss": 5.1722,
|
|
"mean_token_accuracy": 0.20610058009624482,
|
|
"num_tokens": 14491955.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.668806314468384,
|
|
"epoch": 6.851740438332617,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016904937574730062,
|
|
"loss": 5.0328,
|
|
"mean_token_accuracy": 0.21410267055034637,
|
|
"num_tokens": 14501149.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.680093383789062,
|
|
"epoch": 6.856037816931671,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00016875617821747208,
|
|
"loss": 5.1029,
|
|
"mean_token_accuracy": 0.21058415174484252,
|
|
"num_tokens": 14510134.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.618011856079102,
|
|
"epoch": 6.860335195530726,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016846321268168508,
|
|
"loss": 5.0592,
|
|
"mean_token_accuracy": 0.21163852512836456,
|
|
"num_tokens": 14519842.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.696297883987427,
|
|
"epoch": 6.8646325741297805,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016817047977965905,
|
|
"loss": 5.1374,
|
|
"mean_token_accuracy": 0.20659516602754593,
|
|
"num_tokens": 14528790.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.727760457992554,
|
|
"epoch": 6.868929952728836,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0001678779801506058,
|
|
"loss": 5.1577,
|
|
"mean_token_accuracy": 0.19679084718227385,
|
|
"num_tokens": 14537389.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.694173812866211,
|
|
"epoch": 6.8732273313278895,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00016758571443322774,
|
|
"loss": 5.0497,
|
|
"mean_token_accuracy": 0.20700975507497787,
|
|
"num_tokens": 14545095.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.8732273313278895,
|
|
"eval_entropy": 5.551458546707222,
|
|
"eval_loss": 5.957724571228027,
|
|
"eval_mean_token_accuracy": 0.17630530778016593,
|
|
"eval_num_tokens": 14545095.0,
|
|
"eval_runtime": 2.6476,
|
|
"eval_samples_per_second": 1340.484,
|
|
"eval_steps_per_second": 167.702,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.675696992874146,
|
|
"epoch": 6.877524709926945,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001672936832657162,
|
|
"loss": 5.0206,
|
|
"mean_token_accuracy": 0.21922506392002106,
|
|
"num_tokens": 14553910.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.597411251068115,
|
|
"epoch": 6.8818220885259995,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016700188728575047,
|
|
"loss": 5.0363,
|
|
"mean_token_accuracy": 0.21871596127748488,
|
|
"num_tokens": 14562272.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.632660436630249,
|
|
"epoch": 6.886119467125054,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016671032713049655,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.20208823978900908,
|
|
"num_tokens": 14572265.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.6439718246459964,
|
|
"epoch": 6.890416845724109,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00016641900343660515,
|
|
"loss": 5.1181,
|
|
"mean_token_accuracy": 0.20914661139249802,
|
|
"num_tokens": 14580979.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.595297956466675,
|
|
"epoch": 6.894714224323163,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001661279168402107,
|
|
"loss": 5.0463,
|
|
"mean_token_accuracy": 0.21786785274744033,
|
|
"num_tokens": 14589426.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.626606702804565,
|
|
"epoch": 6.899011602922218,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016583706797693008,
|
|
"loss": 5.048,
|
|
"mean_token_accuracy": 0.2119670256972313,
|
|
"num_tokens": 14597457.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.574995374679565,
|
|
"epoch": 6.903308981521272,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016554645748186105,
|
|
"loss": 4.9566,
|
|
"mean_token_accuracy": 0.22303528785705568,
|
|
"num_tokens": 14605921.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.682518720626831,
|
|
"epoch": 6.907606360120327,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00016525608598958063,
|
|
"loss": 5.1372,
|
|
"mean_token_accuracy": 0.20370239317417144,
|
|
"num_tokens": 14615099.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.607616996765136,
|
|
"epoch": 6.911903738719381,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00016496595413414421,
|
|
"loss": 4.9457,
|
|
"mean_token_accuracy": 0.21714029163122178,
|
|
"num_tokens": 14623933.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.725429725646973,
|
|
"epoch": 6.916201117318436,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016467606254908355,
|
|
"loss": 5.1696,
|
|
"mean_token_accuracy": 0.20037865936756133,
|
|
"num_tokens": 14633667.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.664012098312378,
|
|
"epoch": 6.9204984959174904,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016438641186740632,
|
|
"loss": 5.1478,
|
|
"mean_token_accuracy": 0.20746354758739471,
|
|
"num_tokens": 14642392.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.695912599563599,
|
|
"epoch": 6.924795874516545,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00016409700272159371,
|
|
"loss": 5.1438,
|
|
"mean_token_accuracy": 0.20389245301485062,
|
|
"num_tokens": 14652415.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.6891676902771,
|
|
"epoch": 6.9290932531155995,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016380783574359957,
|
|
"loss": 5.1029,
|
|
"mean_token_accuracy": 0.20906483978033066,
|
|
"num_tokens": 14661500.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.764823579788208,
|
|
"epoch": 6.933390631714654,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001635189115648491,
|
|
"loss": 5.1804,
|
|
"mean_token_accuracy": 0.20235525369644164,
|
|
"num_tokens": 14670542.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.729895687103271,
|
|
"epoch": 6.937688010313709,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016323023081623705,
|
|
"loss": 5.1664,
|
|
"mean_token_accuracy": 0.2060927852988243,
|
|
"num_tokens": 14679988.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.617092323303223,
|
|
"epoch": 6.941985388912763,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00016294179412812702,
|
|
"loss": 5.0077,
|
|
"mean_token_accuracy": 0.2138827249407768,
|
|
"num_tokens": 14688035.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.699340581893921,
|
|
"epoch": 6.946282767511818,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00016265360213034923,
|
|
"loss": 5.1041,
|
|
"mean_token_accuracy": 0.2019607186317444,
|
|
"num_tokens": 14696303.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.662295484542847,
|
|
"epoch": 6.950580146110872,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00016236565545220007,
|
|
"loss": 5.1459,
|
|
"mean_token_accuracy": 0.2042476490139961,
|
|
"num_tokens": 14704157.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.67546272277832,
|
|
"epoch": 6.954877524709927,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016207795472243975,
|
|
"loss": 5.1327,
|
|
"mean_token_accuracy": 0.20788997560739517,
|
|
"num_tokens": 14713978.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.62706413269043,
|
|
"epoch": 6.959174903308981,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016179050056929173,
|
|
"loss": 5.121,
|
|
"mean_token_accuracy": 0.21103571951389313,
|
|
"num_tokens": 14724102.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.678141117095947,
|
|
"epoch": 6.963472281908036,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00016150329362044102,
|
|
"loss": 5.0908,
|
|
"mean_token_accuracy": 0.20692331790924073,
|
|
"num_tokens": 14733395.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.658696031570434,
|
|
"epoch": 6.9677696605070905,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016121633450303285,
|
|
"loss": 5.0945,
|
|
"mean_token_accuracy": 0.2098447024822235,
|
|
"num_tokens": 14742395.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.600754499435425,
|
|
"epoch": 6.972067039106145,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016092962384367122,
|
|
"loss": 5.0793,
|
|
"mean_token_accuracy": 0.21118600368499757,
|
|
"num_tokens": 14752004.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.663641881942749,
|
|
"epoch": 6.9763644177051995,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001606431622684176,
|
|
"loss": 5.1268,
|
|
"mean_token_accuracy": 0.196965691447258,
|
|
"num_tokens": 14760766.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.647021627426147,
|
|
"epoch": 6.980661796304254,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016035695040278935,
|
|
"loss": 5.1149,
|
|
"mean_token_accuracy": 0.20644420087337495,
|
|
"num_tokens": 14769994.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.744442510604858,
|
|
"epoch": 6.984959174903309,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016007098887175914,
|
|
"loss": 5.1561,
|
|
"mean_token_accuracy": 0.2044219568371773,
|
|
"num_tokens": 14779339.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.632695484161377,
|
|
"epoch": 6.989256553502363,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015978527829975254,
|
|
"loss": 5.0199,
|
|
"mean_token_accuracy": 0.21479329317808152,
|
|
"num_tokens": 14788258.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.630495309829712,
|
|
"epoch": 6.993553932101419,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015949981931064714,
|
|
"loss": 5.0962,
|
|
"mean_token_accuracy": 0.20408875644207,
|
|
"num_tokens": 14797680.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.607353591918946,
|
|
"epoch": 6.997851310700472,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001592146125277714,
|
|
"loss": 5.0248,
|
|
"mean_token_accuracy": 0.2127462610602379,
|
|
"num_tokens": 14807601.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.694760746426052,
|
|
"epoch": 7.001718951439622,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00015892965857390278,
|
|
"loss": 5.0258,
|
|
"mean_token_accuracy": 0.20738586948977578,
|
|
"num_tokens": 14815802.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.700262022018433,
|
|
"epoch": 7.006016330038676,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015864495807126704,
|
|
"loss": 5.07,
|
|
"mean_token_accuracy": 0.20683721154928209,
|
|
"num_tokens": 14826123.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.59828815460205,
|
|
"epoch": 7.010313708637731,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00015836051164153602,
|
|
"loss": 4.8399,
|
|
"mean_token_accuracy": 0.23073654472827912,
|
|
"num_tokens": 14834332.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.657144546508789,
|
|
"epoch": 7.014611087236785,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015807631990582733,
|
|
"loss": 4.9068,
|
|
"mean_token_accuracy": 0.2214723601937294,
|
|
"num_tokens": 14842283.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.6825852394104,
|
|
"epoch": 7.01890846583584,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00015779238348470192,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.20588721632957457,
|
|
"num_tokens": 14851415.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.664686441421509,
|
|
"epoch": 7.023205844434894,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00015750870299816345,
|
|
"loss": 5.021,
|
|
"mean_token_accuracy": 0.21185164004564286,
|
|
"num_tokens": 14860545.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.674919891357422,
|
|
"epoch": 7.027503223033949,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015722527906565672,
|
|
"loss": 4.9949,
|
|
"mean_token_accuracy": 0.21511755883693695,
|
|
"num_tokens": 14870290.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.633830976486206,
|
|
"epoch": 7.0318006016330035,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00015694211230606647,
|
|
"loss": 4.933,
|
|
"mean_token_accuracy": 0.22148568332195281,
|
|
"num_tokens": 14880151.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.6271501064300535,
|
|
"epoch": 7.036097980232058,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015665920333771564,
|
|
"loss": 5.0076,
|
|
"mean_token_accuracy": 0.20787488669157028,
|
|
"num_tokens": 14890127.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.60055775642395,
|
|
"epoch": 7.0403953588311134,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015637655277836427,
|
|
"loss": 4.9301,
|
|
"mean_token_accuracy": 0.22267776131629943,
|
|
"num_tokens": 14899988.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.7018146991729735,
|
|
"epoch": 7.044692737430168,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001560941612452081,
|
|
"loss": 5.0068,
|
|
"mean_token_accuracy": 0.21334744542837142,
|
|
"num_tokens": 14910598.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.7034670352935795,
|
|
"epoch": 7.0489901160292225,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0001558120293548777,
|
|
"loss": 4.97,
|
|
"mean_token_accuracy": 0.21416361182928084,
|
|
"num_tokens": 14919300.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.680017518997192,
|
|
"epoch": 7.053287494628277,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015553015772343614,
|
|
"loss": 5.0339,
|
|
"mean_token_accuracy": 0.21492578685283661,
|
|
"num_tokens": 14928327.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.667846441268921,
|
|
"epoch": 7.057584873227332,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00015524854696637847,
|
|
"loss": 5.0094,
|
|
"mean_token_accuracy": 0.21392848640680312,
|
|
"num_tokens": 14936591.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.695559692382813,
|
|
"epoch": 7.061882251826386,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015496719769862981,
|
|
"loss": 5.0077,
|
|
"mean_token_accuracy": 0.2085215628147125,
|
|
"num_tokens": 14944940.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.634983015060425,
|
|
"epoch": 7.066179630425441,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00015468611053454478,
|
|
"loss": 4.9568,
|
|
"mean_token_accuracy": 0.22063851356506348,
|
|
"num_tokens": 14953978.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.625253772735595,
|
|
"epoch": 7.070477009024495,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00015440528608790533,
|
|
"loss": 4.9558,
|
|
"mean_token_accuracy": 0.21749907582998276,
|
|
"num_tokens": 14963004.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.636422491073608,
|
|
"epoch": 7.07477438762355,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0001541247249719197,
|
|
"loss": 4.9374,
|
|
"mean_token_accuracy": 0.22119348794221877,
|
|
"num_tokens": 14971239.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.681583547592163,
|
|
"epoch": 7.079071766222604,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.00015384442779922135,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.21435814201831818,
|
|
"num_tokens": 14980049.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.622706317901612,
|
|
"epoch": 7.083369144821659,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015356439518186726,
|
|
"loss": 4.9805,
|
|
"mean_token_accuracy": 0.21802165359258652,
|
|
"num_tokens": 14988430.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.6663610458374025,
|
|
"epoch": 7.0876665234207135,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00015328462773133672,
|
|
"loss": 5.0167,
|
|
"mean_token_accuracy": 0.20751799046993255,
|
|
"num_tokens": 14996921.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.684941959381104,
|
|
"epoch": 7.091963902019768,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00015300512605852977,
|
|
"loss": 5.0838,
|
|
"mean_token_accuracy": 0.2115155816078186,
|
|
"num_tokens": 15006472.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.5628903865814205,
|
|
"epoch": 7.0962612806188226,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001527258907737668,
|
|
"loss": 4.8539,
|
|
"mean_token_accuracy": 0.22981361895799637,
|
|
"num_tokens": 15014863.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.658899307250977,
|
|
"epoch": 7.100558659217877,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00015244692248678586,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.2071695864200592,
|
|
"num_tokens": 15024201.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.601595735549926,
|
|
"epoch": 7.104856037816932,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001521682218067421,
|
|
"loss": 4.9627,
|
|
"mean_token_accuracy": 0.2194338247179985,
|
|
"num_tokens": 15033650.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.537532138824463,
|
|
"epoch": 7.109153416415986,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015188978934220642,
|
|
"loss": 4.8075,
|
|
"mean_token_accuracy": 0.2355663776397705,
|
|
"num_tokens": 15042966.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.634783792495727,
|
|
"epoch": 7.113450795015041,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001516116257011641,
|
|
"loss": 4.9513,
|
|
"mean_token_accuracy": 0.21738143712282182,
|
|
"num_tokens": 15053036.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.671324634552002,
|
|
"epoch": 7.117748173614095,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001513337314910134,
|
|
"loss": 4.9473,
|
|
"mean_token_accuracy": 0.22391628623008727,
|
|
"num_tokens": 15062017.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.7118700504302975,
|
|
"epoch": 7.12204555221315,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015105610731856416,
|
|
"loss": 5.0591,
|
|
"mean_token_accuracy": 0.21066346466541291,
|
|
"num_tokens": 15071625.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.68517804145813,
|
|
"epoch": 7.126342930812204,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015077875379003653,
|
|
"loss": 5.0054,
|
|
"mean_token_accuracy": 0.21334264427423477,
|
|
"num_tokens": 15080911.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.623796558380127,
|
|
"epoch": 7.130640309411259,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015050167151105988,
|
|
"loss": 4.9694,
|
|
"mean_token_accuracy": 0.21878744214773177,
|
|
"num_tokens": 15090772.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.656873035430908,
|
|
"epoch": 7.1349376880103135,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000150224861086671,
|
|
"loss": 5.0532,
|
|
"mean_token_accuracy": 0.20926974415779115,
|
|
"num_tokens": 15100546.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.727634763717651,
|
|
"epoch": 7.139235066609368,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014994832312131332,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.2115994542837143,
|
|
"num_tokens": 15109926.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.537078619003296,
|
|
"epoch": 7.143532445208423,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014967205821883532,
|
|
"loss": 4.8391,
|
|
"mean_token_accuracy": 0.2329632371664047,
|
|
"num_tokens": 15119880.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.6654016971588135,
|
|
"epoch": 7.147829823807477,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000149396066982489,
|
|
"loss": 4.9357,
|
|
"mean_token_accuracy": 0.216637547314167,
|
|
"num_tokens": 15128506.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.643754577636718,
|
|
"epoch": 7.152127202406532,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014912035001492897,
|
|
"loss": 5.0561,
|
|
"mean_token_accuracy": 0.20767711997032165,
|
|
"num_tokens": 15138941.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.635707902908325,
|
|
"epoch": 7.156424581005586,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00014884490791821058,
|
|
"loss": 4.9476,
|
|
"mean_token_accuracy": 0.22126933336257934,
|
|
"num_tokens": 15148070.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.635561275482178,
|
|
"epoch": 7.160721959604641,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014856974129378981,
|
|
"loss": 4.9604,
|
|
"mean_token_accuracy": 0.2243446081876755,
|
|
"num_tokens": 15157789.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.647475337982177,
|
|
"epoch": 7.165019338203695,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001482948507425203,
|
|
"loss": 5.0136,
|
|
"mean_token_accuracy": 0.21071556806564332,
|
|
"num_tokens": 15166759.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.648803949356079,
|
|
"epoch": 7.169316716802751,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00014802023686465314,
|
|
"loss": 5.0411,
|
|
"mean_token_accuracy": 0.21666453182697296,
|
|
"num_tokens": 15175219.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.6548834323883055,
|
|
"epoch": 7.173614095401805,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014774590025983523,
|
|
"loss": 4.8809,
|
|
"mean_token_accuracy": 0.23104362636804582,
|
|
"num_tokens": 15183353.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.6709894180297855,
|
|
"epoch": 7.17791147400086,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014747184152710807,
|
|
"loss": 4.9236,
|
|
"mean_token_accuracy": 0.22337013632059097,
|
|
"num_tokens": 15192164.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.6953291416168215,
|
|
"epoch": 7.182208852599914,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014719806126490658,
|
|
"loss": 5.0565,
|
|
"mean_token_accuracy": 0.21191398948431014,
|
|
"num_tokens": 15202079.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.687911748886108,
|
|
"epoch": 7.186506231198969,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001469245600710573,
|
|
"loss": 4.9753,
|
|
"mean_token_accuracy": 0.21717123538255692,
|
|
"num_tokens": 15211191.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.615052986145019,
|
|
"epoch": 7.1908036097980235,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014665133854277742,
|
|
"loss": 4.929,
|
|
"mean_token_accuracy": 0.22586828023195266,
|
|
"num_tokens": 15220511.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.6136016845703125,
|
|
"epoch": 7.195100988397078,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001463783972766737,
|
|
"loss": 4.9417,
|
|
"mean_token_accuracy": 0.22096556574106216,
|
|
"num_tokens": 15228774.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.7253368377685545,
|
|
"epoch": 7.199398366996133,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001461057368687407,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.20994772911071777,
|
|
"num_tokens": 15237673.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.680148792266846,
|
|
"epoch": 7.203695745595187,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00014583335791435971,
|
|
"loss": 5.0029,
|
|
"mean_token_accuracy": 0.2113779991865158,
|
|
"num_tokens": 15246509.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.678381729125976,
|
|
"epoch": 7.207993124194242,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014556126100829774,
|
|
"loss": 5.0231,
|
|
"mean_token_accuracy": 0.21589482575654984,
|
|
"num_tokens": 15255637.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.656054592132568,
|
|
"epoch": 7.212290502793296,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00014528944674470546,
|
|
"loss": 5.0585,
|
|
"mean_token_accuracy": 0.2071926087141037,
|
|
"num_tokens": 15264923.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.598602151870727,
|
|
"epoch": 7.216587881392351,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001450179157171166,
|
|
"loss": 4.9697,
|
|
"mean_token_accuracy": 0.22057172209024428,
|
|
"num_tokens": 15273183.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.600034189224243,
|
|
"epoch": 7.220885259991405,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014474666851844632,
|
|
"loss": 4.994,
|
|
"mean_token_accuracy": 0.21547668278217316,
|
|
"num_tokens": 15283191.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.674643039703369,
|
|
"epoch": 7.22518263859046,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014447570574099028,
|
|
"loss": 4.983,
|
|
"mean_token_accuracy": 0.21752814799547196,
|
|
"num_tokens": 15292061.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.662658071517944,
|
|
"epoch": 7.229480017189514,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014420502797642283,
|
|
"loss": 4.9993,
|
|
"mean_token_accuracy": 0.2147320255637169,
|
|
"num_tokens": 15300921.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.67653636932373,
|
|
"epoch": 7.233777395788569,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.000143934635815796,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.21466162502765657,
|
|
"num_tokens": 15310371.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.625755405426025,
|
|
"epoch": 7.2380747743876235,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0001436645298495381,
|
|
"loss": 5.072,
|
|
"mean_token_accuracy": 0.1993014082312584,
|
|
"num_tokens": 15318787.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.6988404273986815,
|
|
"epoch": 7.242372152986678,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00014339471066745262,
|
|
"loss": 5.0842,
|
|
"mean_token_accuracy": 0.2053224503993988,
|
|
"num_tokens": 15327431.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.715331649780273,
|
|
"epoch": 7.246669531585733,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000143125178858717,
|
|
"loss": 4.9914,
|
|
"mean_token_accuracy": 0.21752191781997682,
|
|
"num_tokens": 15336376.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.704099225997925,
|
|
"epoch": 7.250966910184787,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00014285593501188083,
|
|
"loss": 5.0522,
|
|
"mean_token_accuracy": 0.2080134630203247,
|
|
"num_tokens": 15346827.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.677044630050659,
|
|
"epoch": 7.255264288783842,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014258697971486492,
|
|
"loss": 5.0229,
|
|
"mean_token_accuracy": 0.20849804133176802,
|
|
"num_tokens": 15355896.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.61549129486084,
|
|
"epoch": 7.259561667382896,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014231831355496045,
|
|
"loss": 4.9975,
|
|
"mean_token_accuracy": 0.21670854687690735,
|
|
"num_tokens": 15365591.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.629289770126343,
|
|
"epoch": 7.263859045981951,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00014204993711882662,
|
|
"loss": 4.9104,
|
|
"mean_token_accuracy": 0.21519649475812913,
|
|
"num_tokens": 15374018.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.58354606628418,
|
|
"epoch": 7.268156424581005,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001417818509924906,
|
|
"loss": 4.9186,
|
|
"mean_token_accuracy": 0.22583243548870086,
|
|
"num_tokens": 15382915.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.645354270935059,
|
|
"epoch": 7.27245380318006,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000141514055761345,
|
|
"loss": 4.9921,
|
|
"mean_token_accuracy": 0.2151130124926567,
|
|
"num_tokens": 15391979.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.59524097442627,
|
|
"epoch": 7.276751181779114,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014124655201014786,
|
|
"loss": 4.9785,
|
|
"mean_token_accuracy": 0.21692243963479996,
|
|
"num_tokens": 15401548.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.685673236846924,
|
|
"epoch": 7.281048560378169,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014097934032302037,
|
|
"loss": 4.9815,
|
|
"mean_token_accuracy": 0.21172913461923598,
|
|
"num_tokens": 15410934.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.6318121433258055,
|
|
"epoch": 7.2853459389772235,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014071242128344593,
|
|
"loss": 4.8806,
|
|
"mean_token_accuracy": 0.22812751084566116,
|
|
"num_tokens": 15420728.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.638222169876099,
|
|
"epoch": 7.289643317576278,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0001404457954742691,
|
|
"loss": 4.9323,
|
|
"mean_token_accuracy": 0.2244494527578354,
|
|
"num_tokens": 15429236.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.662634038925171,
|
|
"epoch": 7.2939406961753335,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00014017946347769423,
|
|
"loss": 4.961,
|
|
"mean_token_accuracy": 0.212709279358387,
|
|
"num_tokens": 15438568.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.6557323932647705,
|
|
"epoch": 7.298238074774388,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00013991342587528377,
|
|
"loss": 4.961,
|
|
"mean_token_accuracy": 0.21428968012332916,
|
|
"num_tokens": 15447166.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.605850410461426,
|
|
"epoch": 7.302535453373443,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00013964768324795752,
|
|
"loss": 4.9203,
|
|
"mean_token_accuracy": 0.22439051270484925,
|
|
"num_tokens": 15456239.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.302535453373443,
|
|
"eval_entropy": 5.4941427256609945,
|
|
"eval_loss": 5.958581447601318,
|
|
"eval_mean_token_accuracy": 0.17654259585112603,
|
|
"eval_num_tokens": 15456239.0,
|
|
"eval_runtime": 2.6386,
|
|
"eval_samples_per_second": 1345.043,
|
|
"eval_steps_per_second": 168.273,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.638356256484985,
|
|
"epoch": 7.306832831972497,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013938223617599124,
|
|
"loss": 5.083,
|
|
"mean_token_accuracy": 0.21253616362810135,
|
|
"num_tokens": 15465336.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.6570042133331295,
|
|
"epoch": 7.311130210571552,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013911708523901514,
|
|
"loss": 4.956,
|
|
"mean_token_accuracy": 0.22150272130966187,
|
|
"num_tokens": 15473856.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.660481023788452,
|
|
"epoch": 7.315427589170606,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013885223101601303,
|
|
"loss": 5.0224,
|
|
"mean_token_accuracy": 0.21711659282445908,
|
|
"num_tokens": 15483009.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.657289981842041,
|
|
"epoch": 7.319724967769661,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013858767408532051,
|
|
"loss": 5.028,
|
|
"mean_token_accuracy": 0.21768376380205154,
|
|
"num_tokens": 15492110.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.707041072845459,
|
|
"epoch": 7.324022346368715,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00013832341502462432,
|
|
"loss": 5.1069,
|
|
"mean_token_accuracy": 0.1980769604444504,
|
|
"num_tokens": 15500782.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.649021291732788,
|
|
"epoch": 7.32831972496777,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013805945441096057,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.2097732573747635,
|
|
"num_tokens": 15509425.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.712623119354248,
|
|
"epoch": 7.332617103566824,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00013779579282071364,
|
|
"loss": 5.0054,
|
|
"mean_token_accuracy": 0.21533941477537155,
|
|
"num_tokens": 15518170.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.729481601715088,
|
|
"epoch": 7.336914482165879,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013753243082961512,
|
|
"loss": 5.0274,
|
|
"mean_token_accuracy": 0.2080045223236084,
|
|
"num_tokens": 15527385.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.6691559791564945,
|
|
"epoch": 7.3412118607649335,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013726936901274246,
|
|
"loss": 5.0226,
|
|
"mean_token_accuracy": 0.21848932951688765,
|
|
"num_tokens": 15536325.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.711063480377197,
|
|
"epoch": 7.345509239363988,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001370066079445174,
|
|
"loss": 5.1097,
|
|
"mean_token_accuracy": 0.19984171390533448,
|
|
"num_tokens": 15545989.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.609355115890503,
|
|
"epoch": 7.349806617963043,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013674414819870502,
|
|
"loss": 4.9635,
|
|
"mean_token_accuracy": 0.22090065181255342,
|
|
"num_tokens": 15555939.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.6611896514892575,
|
|
"epoch": 7.354103996562097,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013648199034841264,
|
|
"loss": 5.0154,
|
|
"mean_token_accuracy": 0.22480156570672988,
|
|
"num_tokens": 15564574.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.68861722946167,
|
|
"epoch": 7.358401375161152,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001362201349660882,
|
|
"loss": 4.9966,
|
|
"mean_token_accuracy": 0.21718013286590576,
|
|
"num_tokens": 15573522.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.711553764343262,
|
|
"epoch": 7.362698753760206,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001359585826235192,
|
|
"loss": 5.0404,
|
|
"mean_token_accuracy": 0.2079370751976967,
|
|
"num_tokens": 15582713.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.67584376335144,
|
|
"epoch": 7.366996132359261,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013569733389183126,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.21447585374116898,
|
|
"num_tokens": 15592667.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.6157654285430905,
|
|
"epoch": 7.371293510958315,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00013543638934148736,
|
|
"loss": 4.9446,
|
|
"mean_token_accuracy": 0.2239904597401619,
|
|
"num_tokens": 15600240.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.636140823364258,
|
|
"epoch": 7.37559088955737,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000135175749542286,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.21127722263336182,
|
|
"num_tokens": 15609711.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.666356992721558,
|
|
"epoch": 7.379888268156424,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001349154150633604,
|
|
"loss": 5.0314,
|
|
"mean_token_accuracy": 0.21117882579565048,
|
|
"num_tokens": 15619060.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.639746856689453,
|
|
"epoch": 7.384185646755479,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000134655386473177,
|
|
"loss": 5.0325,
|
|
"mean_token_accuracy": 0.2090388596057892,
|
|
"num_tokens": 15627953.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.549829483032227,
|
|
"epoch": 7.3884830253545335,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00013439566433953427,
|
|
"loss": 4.7678,
|
|
"mean_token_accuracy": 0.2413141906261444,
|
|
"num_tokens": 15636956.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.642958116531372,
|
|
"epoch": 7.392780403953588,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0001341362492295616,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.21645078361034392,
|
|
"num_tokens": 15646093.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.702984380722046,
|
|
"epoch": 7.397077782552643,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013387714170971776,
|
|
"loss": 5.0573,
|
|
"mean_token_accuracy": 0.21302583068609238,
|
|
"num_tokens": 15655102.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.660951757431031,
|
|
"epoch": 7.401375161151697,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00013361834234579012,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.22242118567228317,
|
|
"num_tokens": 15663989.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.6734425067901615,
|
|
"epoch": 7.405672539750752,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001333598517028931,
|
|
"loss": 5.0033,
|
|
"mean_token_accuracy": 0.2152569606900215,
|
|
"num_tokens": 15673567.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.709622764587403,
|
|
"epoch": 7.409969918349806,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013310167034546688,
|
|
"loss": 5.0133,
|
|
"mean_token_accuracy": 0.21349478662014007,
|
|
"num_tokens": 15683492.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.630057764053345,
|
|
"epoch": 7.414267296948861,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0001328437988372763,
|
|
"loss": 5.0275,
|
|
"mean_token_accuracy": 0.21063388288021087,
|
|
"num_tokens": 15693468.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.687697458267212,
|
|
"epoch": 7.418564675547916,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013258623774140967,
|
|
"loss": 5.0057,
|
|
"mean_token_accuracy": 0.21883064806461333,
|
|
"num_tokens": 15703147.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.664934968948364,
|
|
"epoch": 7.422862054146971,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00013232898762027766,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.2177565187215805,
|
|
"num_tokens": 15711859.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.631044101715088,
|
|
"epoch": 7.427159432746025,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00013207204903561154,
|
|
"loss": 5.0521,
|
|
"mean_token_accuracy": 0.20895788222551345,
|
|
"num_tokens": 15721868.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.60103931427002,
|
|
"epoch": 7.43145681134508,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00013181542254846247,
|
|
"loss": 5.0056,
|
|
"mean_token_accuracy": 0.21752475202083588,
|
|
"num_tokens": 15730913.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.639897584915161,
|
|
"epoch": 7.435754189944134,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001315591087192002,
|
|
"loss": 4.9852,
|
|
"mean_token_accuracy": 0.22172940522432327,
|
|
"num_tokens": 15740108.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.683353042602539,
|
|
"epoch": 7.440051568543189,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00013130310810751162,
|
|
"loss": 5.0153,
|
|
"mean_token_accuracy": 0.21216270327568054,
|
|
"num_tokens": 15749230.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.651005268096924,
|
|
"epoch": 7.4443489471422435,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00013104742127239983,
|
|
"loss": 5.018,
|
|
"mean_token_accuracy": 0.21438120901584626,
|
|
"num_tokens": 15758147.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.749051952362061,
|
|
"epoch": 7.448646325741298,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001307920487721826,
|
|
"loss": 5.0803,
|
|
"mean_token_accuracy": 0.20021321177482604,
|
|
"num_tokens": 15767334.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.620826244354248,
|
|
"epoch": 7.452943704340353,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013053699116449144,
|
|
"loss": 4.904,
|
|
"mean_token_accuracy": 0.22932035624980926,
|
|
"num_tokens": 15776602.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.640384864807129,
|
|
"epoch": 7.457241082939407,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.00013028224900627026,
|
|
"loss": 4.9438,
|
|
"mean_token_accuracy": 0.22566048800945282,
|
|
"num_tokens": 15785336.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.665958595275879,
|
|
"epoch": 7.461538461538462,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013002782285377395,
|
|
"loss": 5.0579,
|
|
"mean_token_accuracy": 0.21310702562332154,
|
|
"num_tokens": 15794226.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.649094343185425,
|
|
"epoch": 7.465835840137516,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001297737132625677,
|
|
"loss": 4.9708,
|
|
"mean_token_accuracy": 0.2187441185116768,
|
|
"num_tokens": 15803456.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.635722017288208,
|
|
"epoch": 7.470133218736571,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00012951992078752528,
|
|
"loss": 5.0142,
|
|
"mean_token_accuracy": 0.21763473451137544,
|
|
"num_tokens": 15812851.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.679235172271729,
|
|
"epoch": 7.474430597335625,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012926644598282798,
|
|
"loss": 4.9824,
|
|
"mean_token_accuracy": 0.21933030784130098,
|
|
"num_tokens": 15821746.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.619796562194824,
|
|
"epoch": 7.47872797593468,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001290132894019634,
|
|
"loss": 4.9608,
|
|
"mean_token_accuracy": 0.21920393258333207,
|
|
"num_tokens": 15830753.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.632902574539185,
|
|
"epoch": 7.483025354533734,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00012876045159772442,
|
|
"loss": 5.0184,
|
|
"mean_token_accuracy": 0.22004060745239257,
|
|
"num_tokens": 15839658.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.64514422416687,
|
|
"epoch": 7.487322733132789,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012850793312220766,
|
|
"loss": 4.9692,
|
|
"mean_token_accuracy": 0.2209884375333786,
|
|
"num_tokens": 15848323.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.672580718994141,
|
|
"epoch": 7.4916201117318435,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00012825573452681266,
|
|
"loss": 5.061,
|
|
"mean_token_accuracy": 0.2101320743560791,
|
|
"num_tokens": 15856967.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.643987369537354,
|
|
"epoch": 7.495917490330898,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012800385636224017,
|
|
"loss": 5.0137,
|
|
"mean_token_accuracy": 0.21777698397636414,
|
|
"num_tokens": 15866339.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.693445825576783,
|
|
"epoch": 7.500214868929953,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012775229917849162,
|
|
"loss": 5.0535,
|
|
"mean_token_accuracy": 0.21730823814868927,
|
|
"num_tokens": 15875768.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.699502182006836,
|
|
"epoch": 7.504512247529007,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012750106352486728,
|
|
"loss": 5.0075,
|
|
"mean_token_accuracy": 0.212035171687603,
|
|
"num_tokens": 15884456.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.617113304138184,
|
|
"epoch": 7.508809626128062,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012725014994996534,
|
|
"loss": 5.0024,
|
|
"mean_token_accuracy": 0.2142792448401451,
|
|
"num_tokens": 15894171.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.672166872024536,
|
|
"epoch": 7.513107004727116,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00012699955900168075,
|
|
"loss": 4.9307,
|
|
"mean_token_accuracy": 0.21813268661499025,
|
|
"num_tokens": 15902394.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.702025985717773,
|
|
"epoch": 7.517404383326171,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012674929122720414,
|
|
"loss": 5.0066,
|
|
"mean_token_accuracy": 0.21679683178663253,
|
|
"num_tokens": 15913151.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.646392059326172,
|
|
"epoch": 7.521701761925225,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001264993471730202,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.22042083740234375,
|
|
"num_tokens": 15921640.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.6383678913116455,
|
|
"epoch": 7.52599914052428,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012624972738490675,
|
|
"loss": 4.9952,
|
|
"mean_token_accuracy": 0.21430415660142899,
|
|
"num_tokens": 15931688.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.651836252212524,
|
|
"epoch": 7.530296519123334,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012600043240793368,
|
|
"loss": 4.9781,
|
|
"mean_token_accuracy": 0.2129224106669426,
|
|
"num_tokens": 15941484.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.6353936195373535,
|
|
"epoch": 7.534593897722389,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00012575146278646175,
|
|
"loss": 5.0521,
|
|
"mean_token_accuracy": 0.21581801772117615,
|
|
"num_tokens": 15950740.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.697710800170898,
|
|
"epoch": 7.5388912763214435,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00012550281906414097,
|
|
"loss": 5.13,
|
|
"mean_token_accuracy": 0.20139155238866807,
|
|
"num_tokens": 15960395.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.637214231491089,
|
|
"epoch": 7.543188654920499,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00012525450178390972,
|
|
"loss": 5.0016,
|
|
"mean_token_accuracy": 0.2177071377635002,
|
|
"num_tokens": 15969685.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.6281530380249025,
|
|
"epoch": 7.547486033519553,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001250065114879939,
|
|
"loss": 4.8765,
|
|
"mean_token_accuracy": 0.22995788007974624,
|
|
"num_tokens": 15977943.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.6874916553497314,
|
|
"epoch": 7.551783412118608,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012475884871790505,
|
|
"loss": 5.0147,
|
|
"mean_token_accuracy": 0.21561941206455232,
|
|
"num_tokens": 15986446.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.659025239944458,
|
|
"epoch": 7.556080790717663,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012451151401443982,
|
|
"loss": 5.0473,
|
|
"mean_token_accuracy": 0.20799486041069032,
|
|
"num_tokens": 15995507.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.651239538192749,
|
|
"epoch": 7.560378169316717,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012426450791767815,
|
|
"loss": 5.0468,
|
|
"mean_token_accuracy": 0.20805774480104447,
|
|
"num_tokens": 16004879.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.548049020767212,
|
|
"epoch": 7.564675547915772,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012401783096698283,
|
|
"loss": 4.8948,
|
|
"mean_token_accuracy": 0.23641139566898345,
|
|
"num_tokens": 16013578.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.643018817901611,
|
|
"epoch": 7.568972926514826,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012377148370099764,
|
|
"loss": 4.9617,
|
|
"mean_token_accuracy": 0.21638682931661607,
|
|
"num_tokens": 16023404.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.633022260665894,
|
|
"epoch": 7.573270305113881,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012352546665764642,
|
|
"loss": 4.9974,
|
|
"mean_token_accuracy": 0.22389217466115952,
|
|
"num_tokens": 16032602.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.675819301605225,
|
|
"epoch": 7.577567683712935,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012327978037413219,
|
|
"loss": 4.9998,
|
|
"mean_token_accuracy": 0.21221993565559388,
|
|
"num_tokens": 16042218.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.578147840499878,
|
|
"epoch": 7.58186506231199,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012303442538693564,
|
|
"loss": 4.892,
|
|
"mean_token_accuracy": 0.22555215805768966,
|
|
"num_tokens": 16050957.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.6742959976196286,
|
|
"epoch": 7.586162440911044,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012278940223181393,
|
|
"loss": 5.0043,
|
|
"mean_token_accuracy": 0.22576287090778352,
|
|
"num_tokens": 16059738.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.686851453781128,
|
|
"epoch": 7.590459819510099,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012254471144379964,
|
|
"loss": 5.0517,
|
|
"mean_token_accuracy": 0.21205877065658568,
|
|
"num_tokens": 16069184.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.723336505889892,
|
|
"epoch": 7.5947571981091535,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00012230035355719968,
|
|
"loss": 5.0553,
|
|
"mean_token_accuracy": 0.20977017730474473,
|
|
"num_tokens": 16078565.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.670467376708984,
|
|
"epoch": 7.599054576708208,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0001220563291055941,
|
|
"loss": 5.0598,
|
|
"mean_token_accuracy": 0.21055052280426026,
|
|
"num_tokens": 16088541.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.680750322341919,
|
|
"epoch": 7.603351955307263,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001218126386218347,
|
|
"loss": 5.0007,
|
|
"mean_token_accuracy": 0.21600914001464844,
|
|
"num_tokens": 16096805.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.652015018463135,
|
|
"epoch": 7.607649333906317,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00012156928263804403,
|
|
"loss": 4.9366,
|
|
"mean_token_accuracy": 0.21461503207683563,
|
|
"num_tokens": 16105863.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.607600355148316,
|
|
"epoch": 7.611946712505372,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001213262616856144,
|
|
"loss": 4.9266,
|
|
"mean_token_accuracy": 0.22360724508762359,
|
|
"num_tokens": 16114715.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.605248308181762,
|
|
"epoch": 7.616244091104426,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012108357629520635,
|
|
"loss": 4.9646,
|
|
"mean_token_accuracy": 0.22289178818464278,
|
|
"num_tokens": 16123373.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.626610660552979,
|
|
"epoch": 7.620541469703481,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012084122699674785,
|
|
"loss": 4.9076,
|
|
"mean_token_accuracy": 0.2272982895374298,
|
|
"num_tokens": 16132900.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.612386178970337,
|
|
"epoch": 7.624838848302535,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00012059921431943278,
|
|
"loss": 4.8607,
|
|
"mean_token_accuracy": 0.23842085748910904,
|
|
"num_tokens": 16141090.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.659826707839966,
|
|
"epoch": 7.62913622690159,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012035753879172026,
|
|
"loss": 4.9858,
|
|
"mean_token_accuracy": 0.21471437364816665,
|
|
"num_tokens": 16149587.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.644644165039063,
|
|
"epoch": 7.633433605500644,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00012011620094133296,
|
|
"loss": 5.0386,
|
|
"mean_token_accuracy": 0.21213797330856324,
|
|
"num_tokens": 16158503.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.549069118499756,
|
|
"epoch": 7.637730984099699,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.00011987520129525622,
|
|
"loss": 4.9126,
|
|
"mean_token_accuracy": 0.2205449178814888,
|
|
"num_tokens": 16167301.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.619760036468506,
|
|
"epoch": 7.6420283626987535,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000119634540379737,
|
|
"loss": 4.931,
|
|
"mean_token_accuracy": 0.21934739202260972,
|
|
"num_tokens": 16176844.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.651407909393311,
|
|
"epoch": 7.646325741297808,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011939421872028262,
|
|
"loss": 5.0277,
|
|
"mean_token_accuracy": 0.21058297902345657,
|
|
"num_tokens": 16186388.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.633527946472168,
|
|
"epoch": 7.650623119896863,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00011915423684165948,
|
|
"loss": 4.9631,
|
|
"mean_token_accuracy": 0.21958723217248916,
|
|
"num_tokens": 16195450.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.689349126815796,
|
|
"epoch": 7.654920498495917,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00011891459526789198,
|
|
"loss": 5.0397,
|
|
"mean_token_accuracy": 0.21145706623792648,
|
|
"num_tokens": 16204778.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.620783948898316,
|
|
"epoch": 7.659217877094972,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001186752945222616,
|
|
"loss": 4.9867,
|
|
"mean_token_accuracy": 0.21506110727787017,
|
|
"num_tokens": 16214574.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.595845413208008,
|
|
"epoch": 7.663515255694026,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011843633512730562,
|
|
"loss": 4.9421,
|
|
"mean_token_accuracy": 0.21796784698963165,
|
|
"num_tokens": 16223164.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.643932819366455,
|
|
"epoch": 7.667812634293082,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00011819771760481576,
|
|
"loss": 4.9951,
|
|
"mean_token_accuracy": 0.21927218437194823,
|
|
"num_tokens": 16232090.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.7140477180480955,
|
|
"epoch": 7.672110012892135,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011795944247583725,
|
|
"loss": 5.0401,
|
|
"mean_token_accuracy": 0.20878540426492692,
|
|
"num_tokens": 16240703.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.616034984588623,
|
|
"epoch": 7.676407391491191,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00011772151026066789,
|
|
"loss": 4.9678,
|
|
"mean_token_accuracy": 0.21974917203187944,
|
|
"num_tokens": 16250274.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.705725479125976,
|
|
"epoch": 7.680704770090245,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011748392147885642,
|
|
"loss": 5.021,
|
|
"mean_token_accuracy": 0.22127360850572586,
|
|
"num_tokens": 16259453.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.632699012756348,
|
|
"epoch": 7.6850021486893,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011724667664920177,
|
|
"loss": 4.9575,
|
|
"mean_token_accuracy": 0.21220399290323258,
|
|
"num_tokens": 16268014.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.632429695129394,
|
|
"epoch": 7.689299527288354,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011700977628975183,
|
|
"loss": 4.965,
|
|
"mean_token_accuracy": 0.2178291544318199,
|
|
"num_tokens": 16276756.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.547591638565064,
|
|
"epoch": 7.693596905887409,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011677322091780243,
|
|
"loss": 4.9209,
|
|
"mean_token_accuracy": 0.22347164899110794,
|
|
"num_tokens": 16286563.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.657753562927246,
|
|
"epoch": 7.6978942844864635,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001165370110498958,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.22838204652070998,
|
|
"num_tokens": 16295501.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.6504607677459715,
|
|
"epoch": 7.702191663085518,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011630114720181989,
|
|
"loss": 4.9691,
|
|
"mean_token_accuracy": 0.21741376519203187,
|
|
"num_tokens": 16303886.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.688564014434815,
|
|
"epoch": 7.706489041684573,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011606562988860711,
|
|
"loss": 5.0432,
|
|
"mean_token_accuracy": 0.21560966819524766,
|
|
"num_tokens": 16312031.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.699381732940674,
|
|
"epoch": 7.710786420283627,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0001158304596245332,
|
|
"loss": 4.9648,
|
|
"mean_token_accuracy": 0.214211168885231,
|
|
"num_tokens": 16320889.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.667723894119263,
|
|
"epoch": 7.715083798882682,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00011559563692311595,
|
|
"loss": 4.9772,
|
|
"mean_token_accuracy": 0.21756573319435119,
|
|
"num_tokens": 16330002.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.690611696243286,
|
|
"epoch": 7.719381177481736,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011536116229711422,
|
|
"loss": 5.057,
|
|
"mean_token_accuracy": 0.21513792127370834,
|
|
"num_tokens": 16339016.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.622929954528809,
|
|
"epoch": 7.723678556080791,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000115127036258527,
|
|
"loss": 4.9542,
|
|
"mean_token_accuracy": 0.2185430869460106,
|
|
"num_tokens": 16347640.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.633456230163574,
|
|
"epoch": 7.727975934679845,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011489325931859185,
|
|
"loss": 4.9782,
|
|
"mean_token_accuracy": 0.22143753468990326,
|
|
"num_tokens": 16356779.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.630218076705932,
|
|
"epoch": 7.7322733132789,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001146598319877843,
|
|
"loss": 4.9142,
|
|
"mean_token_accuracy": 0.22300618439912795,
|
|
"num_tokens": 16365919.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.7322733132789,
|
|
"eval_entropy": 5.534935549572781,
|
|
"eval_loss": 5.947869300842285,
|
|
"eval_mean_token_accuracy": 0.1772870412515896,
|
|
"eval_num_tokens": 16365919.0,
|
|
"eval_runtime": 2.6442,
|
|
"eval_samples_per_second": 1342.197,
|
|
"eval_steps_per_second": 167.916,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.668812799453735,
|
|
"epoch": 7.736570691877954,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011442675477581621,
|
|
"loss": 5.0406,
|
|
"mean_token_accuracy": 0.21208275705575944,
|
|
"num_tokens": 16375671.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.653517198562622,
|
|
"epoch": 7.740868070477009,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0001141940281916352,
|
|
"loss": 4.981,
|
|
"mean_token_accuracy": 0.21639966517686843,
|
|
"num_tokens": 16384781.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.674461936950683,
|
|
"epoch": 7.7451654490760635,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00011396165274342304,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.21114646643400192,
|
|
"num_tokens": 16393136.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.676620435714722,
|
|
"epoch": 7.749462827675118,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00011372962893859471,
|
|
"loss": 4.9632,
|
|
"mean_token_accuracy": 0.2180955022573471,
|
|
"num_tokens": 16401089.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.620439577102661,
|
|
"epoch": 7.753760206274173,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011349795728379759,
|
|
"loss": 4.9911,
|
|
"mean_token_accuracy": 0.22103059887886048,
|
|
"num_tokens": 16409072.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.573658561706543,
|
|
"epoch": 7.758057584873227,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011326663828491,
|
|
"loss": 4.8772,
|
|
"mean_token_accuracy": 0.23123527318239212,
|
|
"num_tokens": 16417852.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.665076637268067,
|
|
"epoch": 7.762354963472282,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011303567244704015,
|
|
"loss": 4.9924,
|
|
"mean_token_accuracy": 0.2189505562186241,
|
|
"num_tokens": 16427167.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.676531267166138,
|
|
"epoch": 7.766652342071336,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011280506027452502,
|
|
"loss": 5.0617,
|
|
"mean_token_accuracy": 0.2123047098517418,
|
|
"num_tokens": 16436424.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.671193075180054,
|
|
"epoch": 7.770949720670391,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001125748022709295,
|
|
"loss": 5.0211,
|
|
"mean_token_accuracy": 0.21549418419599534,
|
|
"num_tokens": 16445873.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.6975624561309814,
|
|
"epoch": 7.775247099269445,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011234489893904509,
|
|
"loss": 5.0741,
|
|
"mean_token_accuracy": 0.20923960506916045,
|
|
"num_tokens": 16455541.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.737305212020874,
|
|
"epoch": 7.7795444778685,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011211535078088869,
|
|
"loss": 5.0489,
|
|
"mean_token_accuracy": 0.21230971813201904,
|
|
"num_tokens": 16464740.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.664432954788208,
|
|
"epoch": 7.783841856467554,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011188615829770171,
|
|
"loss": 4.9593,
|
|
"mean_token_accuracy": 0.21703123599290847,
|
|
"num_tokens": 16473234.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.692548131942749,
|
|
"epoch": 7.788139235066609,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011165732198994905,
|
|
"loss": 5.0134,
|
|
"mean_token_accuracy": 0.2101627841591835,
|
|
"num_tokens": 16482037.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.6287736892700195,
|
|
"epoch": 7.792436613665664,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011142884235731756,
|
|
"loss": 5.0546,
|
|
"mean_token_accuracy": 0.21271848529577256,
|
|
"num_tokens": 16490844.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.589935302734375,
|
|
"epoch": 7.796733992264718,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00011120071989871564,
|
|
"loss": 4.9344,
|
|
"mean_token_accuracy": 0.21939936578273772,
|
|
"num_tokens": 16499674.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.745057106018066,
|
|
"epoch": 7.8010313708637735,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011097295511227134,
|
|
"loss": 5.1101,
|
|
"mean_token_accuracy": 0.2077337846159935,
|
|
"num_tokens": 16509907.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.63685359954834,
|
|
"epoch": 7.805328749462827,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001107455484953321,
|
|
"loss": 4.9712,
|
|
"mean_token_accuracy": 0.2232545092701912,
|
|
"num_tokens": 16519134.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.647802877426147,
|
|
"epoch": 7.809626128061883,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00011051850054446306,
|
|
"loss": 5.0179,
|
|
"mean_token_accuracy": 0.21318460404872894,
|
|
"num_tokens": 16527969.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.619999504089355,
|
|
"epoch": 7.813923506660937,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00011029181175544603,
|
|
"loss": 4.9352,
|
|
"mean_token_accuracy": 0.21495155841112137,
|
|
"num_tokens": 16536097.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.6448108673095705,
|
|
"epoch": 7.818220885259992,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011006548262327884,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.2173166811466217,
|
|
"num_tokens": 16545249.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.640761995315552,
|
|
"epoch": 7.822518263859046,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001098395136421739,
|
|
"loss": 5.064,
|
|
"mean_token_accuracy": 0.21515174061059952,
|
|
"num_tokens": 16553844.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.699404573440551,
|
|
"epoch": 7.826815642458101,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00010961390530555712,
|
|
"loss": 5.06,
|
|
"mean_token_accuracy": 0.21205902248620986,
|
|
"num_tokens": 16563561.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.580752992630005,
|
|
"epoch": 7.831113021057155,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010938865810606682,
|
|
"loss": 4.8386,
|
|
"mean_token_accuracy": 0.2306146040558815,
|
|
"num_tokens": 16571291.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.67677640914917,
|
|
"epoch": 7.83541039965621,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010916377253555293,
|
|
"loss": 5.0127,
|
|
"mean_token_accuracy": 0.218289814889431,
|
|
"num_tokens": 16580117.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.60970892906189,
|
|
"epoch": 7.839707778255264,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010893924908507573,
|
|
"loss": 4.9395,
|
|
"mean_token_accuracy": 0.2246902823448181,
|
|
"num_tokens": 16588118.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.600351572036743,
|
|
"epoch": 7.844005156854319,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001087150882449046,
|
|
"loss": 5.0042,
|
|
"mean_token_accuracy": 0.22809237539768218,
|
|
"num_tokens": 16598271.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.676624917984009,
|
|
"epoch": 7.8483025354533735,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010849129050451717,
|
|
"loss": 4.9584,
|
|
"mean_token_accuracy": 0.2159911200404167,
|
|
"num_tokens": 16606026.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.68514666557312,
|
|
"epoch": 7.852599914052428,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010826785635259842,
|
|
"loss": 5.042,
|
|
"mean_token_accuracy": 0.21939123123884202,
|
|
"num_tokens": 16614775.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.74629430770874,
|
|
"epoch": 7.856897292651483,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00010804478627703903,
|
|
"loss": 5.0736,
|
|
"mean_token_accuracy": 0.21156911700963973,
|
|
"num_tokens": 16623659.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.689434576034546,
|
|
"epoch": 7.861194671250537,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010782208076493508,
|
|
"loss": 5.0879,
|
|
"mean_token_accuracy": 0.20850563943386077,
|
|
"num_tokens": 16633336.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.635262060165405,
|
|
"epoch": 7.865492049849592,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00010759974030258621,
|
|
"loss": 4.9679,
|
|
"mean_token_accuracy": 0.22224036008119583,
|
|
"num_tokens": 16642327.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.6725990772247314,
|
|
"epoch": 7.869789428448646,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010737776537549531,
|
|
"loss": 5.0292,
|
|
"mean_token_accuracy": 0.21729173511266708,
|
|
"num_tokens": 16651498.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.6510590553283695,
|
|
"epoch": 7.874086807047701,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00010715615646836679,
|
|
"loss": 4.928,
|
|
"mean_token_accuracy": 0.22129568755626677,
|
|
"num_tokens": 16659611.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.609929895401001,
|
|
"epoch": 7.878384185646755,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010693491406510585,
|
|
"loss": 4.9637,
|
|
"mean_token_accuracy": 0.222550268471241,
|
|
"num_tokens": 16668822.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.650983428955078,
|
|
"epoch": 7.88268156424581,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010671403864881757,
|
|
"loss": 4.9625,
|
|
"mean_token_accuracy": 0.22100123763084412,
|
|
"num_tokens": 16677857.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.578815698623657,
|
|
"epoch": 7.8869789428448644,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010649353070180562,
|
|
"loss": 4.8894,
|
|
"mean_token_accuracy": 0.22601640969514847,
|
|
"num_tokens": 16686721.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.609799146652222,
|
|
"epoch": 7.891276321443919,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010627339070557118,
|
|
"loss": 4.9175,
|
|
"mean_token_accuracy": 0.2248716503381729,
|
|
"num_tokens": 16696380.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.56900987625122,
|
|
"epoch": 7.8955737000429735,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00010605361914081194,
|
|
"loss": 5.0087,
|
|
"mean_token_accuracy": 0.2162453293800354,
|
|
"num_tokens": 16705703.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.617312622070313,
|
|
"epoch": 7.899871078642028,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010583421648742125,
|
|
"loss": 4.9364,
|
|
"mean_token_accuracy": 0.2224879667162895,
|
|
"num_tokens": 16714626.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.621264696121216,
|
|
"epoch": 7.904168457241083,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.00010561518322448673,
|
|
"loss": 4.9678,
|
|
"mean_token_accuracy": 0.223540760576725,
|
|
"num_tokens": 16723227.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.6278337955474855,
|
|
"epoch": 7.908465835840137,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00010539651983028955,
|
|
"loss": 5.0079,
|
|
"mean_token_accuracy": 0.21669176667928697,
|
|
"num_tokens": 16731531.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.621719694137573,
|
|
"epoch": 7.912763214439192,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001051782267823031,
|
|
"loss": 5.0188,
|
|
"mean_token_accuracy": 0.21929732859134674,
|
|
"num_tokens": 16741018.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.650992202758789,
|
|
"epoch": 7.917060593038247,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010496030455719225,
|
|
"loss": 5.063,
|
|
"mean_token_accuracy": 0.2096100389957428,
|
|
"num_tokens": 16750997.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.630368900299072,
|
|
"epoch": 7.921357971637301,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00010474275363081193,
|
|
"loss": 4.9596,
|
|
"mean_token_accuracy": 0.22122804969549179,
|
|
"num_tokens": 16760238.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.6477875232696535,
|
|
"epoch": 7.925655350236356,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001045255744782064,
|
|
"loss": 5.0601,
|
|
"mean_token_accuracy": 0.21104993671178818,
|
|
"num_tokens": 16770595.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.646652793884277,
|
|
"epoch": 7.92995272883541,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010430876757360817,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.232233165204525,
|
|
"num_tokens": 16779069.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.602943658828735,
|
|
"epoch": 7.934250107434465,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010409233339043694,
|
|
"loss": 4.9435,
|
|
"mean_token_accuracy": 0.22062616050243378,
|
|
"num_tokens": 16787853.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.6617461204528805,
|
|
"epoch": 7.93854748603352,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010387627240129838,
|
|
"loss": 4.9733,
|
|
"mean_token_accuracy": 0.21271925866603852,
|
|
"num_tokens": 16796069.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.642918729782105,
|
|
"epoch": 7.942844864632574,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010366058507798326,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.21031410992145538,
|
|
"num_tokens": 16805565.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.676023149490357,
|
|
"epoch": 7.947142243231629,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010344527189146655,
|
|
"loss": 5.0114,
|
|
"mean_token_accuracy": 0.20802072435617447,
|
|
"num_tokens": 16814863.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.652223873138428,
|
|
"epoch": 7.9514396218306835,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010323033331190626,
|
|
"loss": 4.9924,
|
|
"mean_token_accuracy": 0.21847451478242874,
|
|
"num_tokens": 16823539.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.663642549514771,
|
|
"epoch": 7.955737000429738,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010301576980864228,
|
|
"loss": 4.9342,
|
|
"mean_token_accuracy": 0.2202010467648506,
|
|
"num_tokens": 16831172.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.6208044528961185,
|
|
"epoch": 7.960034379028793,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00010280158185019547,
|
|
"loss": 4.9905,
|
|
"mean_token_accuracy": 0.2121267333626747,
|
|
"num_tokens": 16840952.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.633737993240357,
|
|
"epoch": 7.964331757627847,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010258776990426686,
|
|
"loss": 5.0194,
|
|
"mean_token_accuracy": 0.21141452193260193,
|
|
"num_tokens": 16849821.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.609160566329956,
|
|
"epoch": 7.968629136226902,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010237433443773612,
|
|
"loss": 4.9673,
|
|
"mean_token_accuracy": 0.21956382542848588,
|
|
"num_tokens": 16858405.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.529933071136474,
|
|
"epoch": 7.972926514825956,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00010216127591666115,
|
|
"loss": 4.8739,
|
|
"mean_token_accuracy": 0.22767941206693648,
|
|
"num_tokens": 16867542.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.552528238296508,
|
|
"epoch": 7.977223893425011,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00010194859480627648,
|
|
"loss": 4.9286,
|
|
"mean_token_accuracy": 0.2205257922410965,
|
|
"num_tokens": 16875984.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.63018479347229,
|
|
"epoch": 7.981521272024065,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010173629157099279,
|
|
"loss": 5.0122,
|
|
"mean_token_accuracy": 0.20925173312425613,
|
|
"num_tokens": 16885268.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.641224002838134,
|
|
"epoch": 7.98581865062312,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010152436667439537,
|
|
"loss": 5.0227,
|
|
"mean_token_accuracy": 0.21047758758068086,
|
|
"num_tokens": 16894637.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.692606687545776,
|
|
"epoch": 7.9901160292221745,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010131282057924345,
|
|
"loss": 5.0865,
|
|
"mean_token_accuracy": 0.20759778022766112,
|
|
"num_tokens": 16904538.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.683877658843994,
|
|
"epoch": 7.994413407821229,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010110165374746924,
|
|
"loss": 5.0623,
|
|
"mean_token_accuracy": 0.21000971794128417,
|
|
"num_tokens": 16914254.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.707946920394898,
|
|
"epoch": 7.9987107864202835,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00010089086664017674,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.2135366305708885,
|
|
"num_tokens": 16924936.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.68029769261678,
|
|
"epoch": 8.002578427159433,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010068045971764067,
|
|
"loss": 4.9981,
|
|
"mean_token_accuracy": 0.2181948588954078,
|
|
"num_tokens": 16932966.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.6510279178619385,
|
|
"epoch": 8.006875805758487,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010047043343930561,
|
|
"loss": 4.8993,
|
|
"mean_token_accuracy": 0.23105971664190292,
|
|
"num_tokens": 16941814.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.617275953292847,
|
|
"epoch": 8.011173184357542,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00010026078826378502,
|
|
"loss": 5.0001,
|
|
"mean_token_accuracy": 0.22531700879335403,
|
|
"num_tokens": 16950995.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.614319324493408,
|
|
"epoch": 8.015470562956596,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00010005152464886031,
|
|
"loss": 4.8746,
|
|
"mean_token_accuracy": 0.22857287526130676,
|
|
"num_tokens": 16959599.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.682064867019653,
|
|
"epoch": 8.019767941555651,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.984264305147941e-05,
|
|
"loss": 4.995,
|
|
"mean_token_accuracy": 0.2170554667711258,
|
|
"num_tokens": 16969425.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.650733280181885,
|
|
"epoch": 8.024065320154705,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.963414392775627e-05,
|
|
"loss": 4.9007,
|
|
"mean_token_accuracy": 0.2346941664814949,
|
|
"num_tokens": 16977907.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.635744714736939,
|
|
"epoch": 8.02836269875376,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.942602773296971e-05,
|
|
"loss": 4.979,
|
|
"mean_token_accuracy": 0.21409488767385482,
|
|
"num_tokens": 16987172.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.657286596298218,
|
|
"epoch": 8.032660077352816,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.921829492156223e-05,
|
|
"loss": 4.9547,
|
|
"mean_token_accuracy": 0.22037395536899568,
|
|
"num_tokens": 16997745.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.654837942123413,
|
|
"epoch": 8.03695745595187,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 9.901094594713933e-05,
|
|
"loss": 4.9705,
|
|
"mean_token_accuracy": 0.21459636390209197,
|
|
"num_tokens": 17007978.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.607391595840454,
|
|
"epoch": 8.041254834550925,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.88039812624682e-05,
|
|
"loss": 4.8803,
|
|
"mean_token_accuracy": 0.22962852269411088,
|
|
"num_tokens": 17016337.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.5807445526123045,
|
|
"epoch": 8.045552213149978,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.859740131947715e-05,
|
|
"loss": 4.7867,
|
|
"mean_token_accuracy": 0.2389984279870987,
|
|
"num_tokens": 17025182.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.6539867401123045,
|
|
"epoch": 8.049849591749034,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.839120656925407e-05,
|
|
"loss": 4.8828,
|
|
"mean_token_accuracy": 0.22960586696863175,
|
|
"num_tokens": 17034005.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.620291662216187,
|
|
"epoch": 8.054146970348087,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.818539746204588e-05,
|
|
"loss": 4.9392,
|
|
"mean_token_accuracy": 0.21822922080755233,
|
|
"num_tokens": 17043303.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.663563203811646,
|
|
"epoch": 8.058444348947143,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 9.797997444725745e-05,
|
|
"loss": 4.9524,
|
|
"mean_token_accuracy": 0.21539316177368165,
|
|
"num_tokens": 17051382.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.625338220596314,
|
|
"epoch": 8.062741727546197,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.77749379734506e-05,
|
|
"loss": 4.9425,
|
|
"mean_token_accuracy": 0.2219965487718582,
|
|
"num_tokens": 17061261.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.608913278579712,
|
|
"epoch": 8.067039106145252,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.757028848834293e-05,
|
|
"loss": 4.9344,
|
|
"mean_token_accuracy": 0.22068460583686828,
|
|
"num_tokens": 17071157.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.650302457809448,
|
|
"epoch": 8.071336484744306,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.736602643880712e-05,
|
|
"loss": 4.9388,
|
|
"mean_token_accuracy": 0.22174968868494033,
|
|
"num_tokens": 17079642.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.611996173858643,
|
|
"epoch": 8.075633863343361,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.716215227086997e-05,
|
|
"loss": 4.8859,
|
|
"mean_token_accuracy": 0.2256158247590065,
|
|
"num_tokens": 17089600.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.606341028213501,
|
|
"epoch": 8.079931241942415,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.695866642971098e-05,
|
|
"loss": 4.8271,
|
|
"mean_token_accuracy": 0.23555005788803102,
|
|
"num_tokens": 17098344.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.599076986312866,
|
|
"epoch": 8.08422862054147,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 9.67555693596621e-05,
|
|
"loss": 4.8459,
|
|
"mean_token_accuracy": 0.2208498477935791,
|
|
"num_tokens": 17106719.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.6819010257720945,
|
|
"epoch": 8.088525999140524,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.655286150420595e-05,
|
|
"loss": 4.9496,
|
|
"mean_token_accuracy": 0.22124833315610887,
|
|
"num_tokens": 17116177.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.643929147720337,
|
|
"epoch": 8.09282337773958,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.635054330597565e-05,
|
|
"loss": 4.9443,
|
|
"mean_token_accuracy": 0.22369855493307114,
|
|
"num_tokens": 17126076.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.697405433654785,
|
|
"epoch": 8.097120756338633,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.614861520675322e-05,
|
|
"loss": 4.9772,
|
|
"mean_token_accuracy": 0.2200745314359665,
|
|
"num_tokens": 17136229.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.667930793762207,
|
|
"epoch": 8.101418134937688,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.594707764746881e-05,
|
|
"loss": 4.9478,
|
|
"mean_token_accuracy": 0.21915210783481598,
|
|
"num_tokens": 17144966.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.646083974838257,
|
|
"epoch": 8.105715513536742,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.57459310682e-05,
|
|
"loss": 4.8536,
|
|
"mean_token_accuracy": 0.23010584115982055,
|
|
"num_tokens": 17153563.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.703075838088989,
|
|
"epoch": 8.110012892135797,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.554517590817055e-05,
|
|
"loss": 5.0076,
|
|
"mean_token_accuracy": 0.21440771371126174,
|
|
"num_tokens": 17162696.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.559389448165893,
|
|
"epoch": 8.114310270734851,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.534481260574944e-05,
|
|
"loss": 4.8596,
|
|
"mean_token_accuracy": 0.22879299968481065,
|
|
"num_tokens": 17171779.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.568219137191773,
|
|
"epoch": 8.118607649333907,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.514484159844997e-05,
|
|
"loss": 4.897,
|
|
"mean_token_accuracy": 0.22422071546316147,
|
|
"num_tokens": 17180479.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.644136714935303,
|
|
"epoch": 8.12290502793296,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 9.494526332292899e-05,
|
|
"loss": 4.9533,
|
|
"mean_token_accuracy": 0.21535244584083557,
|
|
"num_tokens": 17189031.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.686622285842896,
|
|
"epoch": 8.127202406532016,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.47460782149857e-05,
|
|
"loss": 4.9865,
|
|
"mean_token_accuracy": 0.21990122348070146,
|
|
"num_tokens": 17197497.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.67869725227356,
|
|
"epoch": 8.13149978513107,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.454728670956073e-05,
|
|
"loss": 4.9795,
|
|
"mean_token_accuracy": 0.21501413136720657,
|
|
"num_tokens": 17206663.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.611683940887451,
|
|
"epoch": 8.135797163730125,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.43488892407352e-05,
|
|
"loss": 4.8931,
|
|
"mean_token_accuracy": 0.22541621774435044,
|
|
"num_tokens": 17215578.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.605654525756836,
|
|
"epoch": 8.140094542329178,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.415088624172997e-05,
|
|
"loss": 4.8769,
|
|
"mean_token_accuracy": 0.2329181909561157,
|
|
"num_tokens": 17225533.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.487248229980469,
|
|
"epoch": 8.144391920928234,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 9.395327814490439e-05,
|
|
"loss": 4.7007,
|
|
"mean_token_accuracy": 0.25239962339401245,
|
|
"num_tokens": 17234408.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.607541131973266,
|
|
"epoch": 8.148689299527287,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.375606538175566e-05,
|
|
"loss": 4.9371,
|
|
"mean_token_accuracy": 0.2177738606929779,
|
|
"num_tokens": 17243346.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.636883306503296,
|
|
"epoch": 8.152986678126343,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.35592483829175e-05,
|
|
"loss": 4.9123,
|
|
"mean_token_accuracy": 0.22153226733207704,
|
|
"num_tokens": 17253112.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.546134662628174,
|
|
"epoch": 8.157284056725398,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.336282757815964e-05,
|
|
"loss": 4.8491,
|
|
"mean_token_accuracy": 0.22613907158374785,
|
|
"num_tokens": 17261857.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.650029277801513,
|
|
"epoch": 8.161581435324452,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.316680339638664e-05,
|
|
"loss": 4.9112,
|
|
"mean_token_accuracy": 0.2216758608818054,
|
|
"num_tokens": 17269671.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 8.161581435324452,
|
|
"eval_entropy": 5.515215150407843,
|
|
"eval_loss": 5.9452900886535645,
|
|
"eval_mean_token_accuracy": 0.1774668138105053,
|
|
"eval_num_tokens": 17269671.0,
|
|
"eval_runtime": 2.6445,
|
|
"eval_samples_per_second": 1342.052,
|
|
"eval_steps_per_second": 167.898,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.660151481628418,
|
|
"epoch": 8.165878813923507,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.297117626563687e-05,
|
|
"loss": 4.8899,
|
|
"mean_token_accuracy": 0.23193657249212266,
|
|
"num_tokens": 17278039.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.674979114532471,
|
|
"epoch": 8.170176192522561,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.27759466130818e-05,
|
|
"loss": 4.9134,
|
|
"mean_token_accuracy": 0.21830762475728988,
|
|
"num_tokens": 17288324.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.660159158706665,
|
|
"epoch": 8.174473571121617,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.25811148650251e-05,
|
|
"loss": 4.9596,
|
|
"mean_token_accuracy": 0.21940231919288636,
|
|
"num_tokens": 17297101.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.616815614700317,
|
|
"epoch": 8.17877094972067,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.238668144690133e-05,
|
|
"loss": 4.9843,
|
|
"mean_token_accuracy": 0.21658325791358948,
|
|
"num_tokens": 17306124.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.669058704376221,
|
|
"epoch": 8.183068328319726,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.219264678327527e-05,
|
|
"loss": 4.9066,
|
|
"mean_token_accuracy": 0.22411565929651261,
|
|
"num_tokens": 17315010.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.685875988006591,
|
|
"epoch": 8.18736570691878,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 9.199901129784121e-05,
|
|
"loss": 5.0109,
|
|
"mean_token_accuracy": 0.21452863961458207,
|
|
"num_tokens": 17324203.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.661561679840088,
|
|
"epoch": 8.191663085517835,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 9.180577541342164e-05,
|
|
"loss": 4.9399,
|
|
"mean_token_accuracy": 0.22336852848529815,
|
|
"num_tokens": 17333180.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.6258776664733885,
|
|
"epoch": 8.195960464116888,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.161293955196648e-05,
|
|
"loss": 4.9043,
|
|
"mean_token_accuracy": 0.22140600085258483,
|
|
"num_tokens": 17342446.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.696708297729492,
|
|
"epoch": 8.200257842715944,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.142050413455214e-05,
|
|
"loss": 4.942,
|
|
"mean_token_accuracy": 0.21926558911800384,
|
|
"num_tokens": 17350648.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.665114688873291,
|
|
"epoch": 8.204555221314997,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.12284695813807e-05,
|
|
"loss": 4.9773,
|
|
"mean_token_accuracy": 0.21348939090967178,
|
|
"num_tokens": 17360132.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.622413063049317,
|
|
"epoch": 8.208852599914053,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 9.103683631177878e-05,
|
|
"loss": 4.9198,
|
|
"mean_token_accuracy": 0.22215647846460343,
|
|
"num_tokens": 17369088.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.641799116134644,
|
|
"epoch": 8.213149978513107,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.084560474419701e-05,
|
|
"loss": 4.8781,
|
|
"mean_token_accuracy": 0.2285885691642761,
|
|
"num_tokens": 17378385.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.693100357055664,
|
|
"epoch": 8.217447357112162,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.065477529620852e-05,
|
|
"loss": 4.9567,
|
|
"mean_token_accuracy": 0.21662483513355255,
|
|
"num_tokens": 17388084.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.618384647369385,
|
|
"epoch": 8.221744735711216,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 9.046434838450868e-05,
|
|
"loss": 4.8459,
|
|
"mean_token_accuracy": 0.22226353883743286,
|
|
"num_tokens": 17397754.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.675736331939698,
|
|
"epoch": 8.226042114310271,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.027432442491369e-05,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.22039028257131577,
|
|
"num_tokens": 17406447.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.6400031566619875,
|
|
"epoch": 8.230339492909325,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.008470383235991e-05,
|
|
"loss": 4.9395,
|
|
"mean_token_accuracy": 0.2220180869102478,
|
|
"num_tokens": 17416056.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.675000143051148,
|
|
"epoch": 8.23463687150838,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.989548702090295e-05,
|
|
"loss": 5.059,
|
|
"mean_token_accuracy": 0.2099338576197624,
|
|
"num_tokens": 17426497.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.663261795043946,
|
|
"epoch": 8.238934250107434,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.970667440371676e-05,
|
|
"loss": 4.8994,
|
|
"mean_token_accuracy": 0.2270934462547302,
|
|
"num_tokens": 17435889.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.63062744140625,
|
|
"epoch": 8.24323162870649,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.951826639309257e-05,
|
|
"loss": 4.8864,
|
|
"mean_token_accuracy": 0.2325778305530548,
|
|
"num_tokens": 17444722.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.677574920654297,
|
|
"epoch": 8.247529007305543,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.933026340043811e-05,
|
|
"loss": 4.9391,
|
|
"mean_token_accuracy": 0.2208372414112091,
|
|
"num_tokens": 17453349.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.683074998855591,
|
|
"epoch": 8.251826385904598,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.914266583627684e-05,
|
|
"loss": 4.9517,
|
|
"mean_token_accuracy": 0.21906218379735948,
|
|
"num_tokens": 17462000.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.607191514968872,
|
|
"epoch": 8.256123764503652,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.89554741102469e-05,
|
|
"loss": 4.9211,
|
|
"mean_token_accuracy": 0.2319620281457901,
|
|
"num_tokens": 17470530.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.659463834762573,
|
|
"epoch": 8.260421143102707,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.876868863110013e-05,
|
|
"loss": 4.9103,
|
|
"mean_token_accuracy": 0.22219522893428803,
|
|
"num_tokens": 17478741.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.632245445251465,
|
|
"epoch": 8.264718521701761,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.858230980670134e-05,
|
|
"loss": 4.9467,
|
|
"mean_token_accuracy": 0.2161929041147232,
|
|
"num_tokens": 17488401.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.654288721084595,
|
|
"epoch": 8.269015900300817,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.839633804402747e-05,
|
|
"loss": 4.967,
|
|
"mean_token_accuracy": 0.22239744067192077,
|
|
"num_tokens": 17497840.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.655004215240479,
|
|
"epoch": 8.27331327889987,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.821077374916647e-05,
|
|
"loss": 4.9939,
|
|
"mean_token_accuracy": 0.21919782012701033,
|
|
"num_tokens": 17508426.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.603233385086059,
|
|
"epoch": 8.277610657498926,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 8.802561732731654e-05,
|
|
"loss": 4.8916,
|
|
"mean_token_accuracy": 0.21907159090042114,
|
|
"num_tokens": 17517546.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.677417325973511,
|
|
"epoch": 8.281908036097981,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.784086918278534e-05,
|
|
"loss": 4.9715,
|
|
"mean_token_accuracy": 0.2278150200843811,
|
|
"num_tokens": 17527156.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.6728129386901855,
|
|
"epoch": 8.286205414697035,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.765652971898908e-05,
|
|
"loss": 4.9511,
|
|
"mean_token_accuracy": 0.21434662193059922,
|
|
"num_tokens": 17536131.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.613059711456299,
|
|
"epoch": 8.29050279329609,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.747259933845134e-05,
|
|
"loss": 4.8368,
|
|
"mean_token_accuracy": 0.23268184959888458,
|
|
"num_tokens": 17544950.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.700673627853393,
|
|
"epoch": 8.294800171895144,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.728907844280254e-05,
|
|
"loss": 5.0435,
|
|
"mean_token_accuracy": 0.21826354563236236,
|
|
"num_tokens": 17554375.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.572335720062256,
|
|
"epoch": 8.2990975504942,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.710596743277901e-05,
|
|
"loss": 4.8448,
|
|
"mean_token_accuracy": 0.23478659391403198,
|
|
"num_tokens": 17562979.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.630917024612427,
|
|
"epoch": 8.303394929093253,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.692326670822207e-05,
|
|
"loss": 4.9183,
|
|
"mean_token_accuracy": 0.21867713183164597,
|
|
"num_tokens": 17571728.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.65310230255127,
|
|
"epoch": 8.307692307692308,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.6740976668077e-05,
|
|
"loss": 4.947,
|
|
"mean_token_accuracy": 0.2193886548280716,
|
|
"num_tokens": 17581074.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.699738645553589,
|
|
"epoch": 8.311989686291362,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.655909771039239e-05,
|
|
"loss": 5.0259,
|
|
"mean_token_accuracy": 0.2093302920460701,
|
|
"num_tokens": 17590656.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.705768966674805,
|
|
"epoch": 8.316287064890417,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 8.637763023231922e-05,
|
|
"loss": 4.9151,
|
|
"mean_token_accuracy": 0.2246680364012718,
|
|
"num_tokens": 17598967.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.652606391906739,
|
|
"epoch": 8.320584443489471,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 8.619657463010987e-05,
|
|
"loss": 4.9459,
|
|
"mean_token_accuracy": 0.22285015881061554,
|
|
"num_tokens": 17608247.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.561486053466797,
|
|
"epoch": 8.324881822088527,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.60159312991175e-05,
|
|
"loss": 4.8614,
|
|
"mean_token_accuracy": 0.23390029221773148,
|
|
"num_tokens": 17616805.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.6130842685699465,
|
|
"epoch": 8.32917920068758,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.583570063379487e-05,
|
|
"loss": 4.8475,
|
|
"mean_token_accuracy": 0.23111422508955,
|
|
"num_tokens": 17625372.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.629535245895386,
|
|
"epoch": 8.333476579286636,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.565588302769374e-05,
|
|
"loss": 4.9635,
|
|
"mean_token_accuracy": 0.21621522158384324,
|
|
"num_tokens": 17633586.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.592336845397949,
|
|
"epoch": 8.33777395788569,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.54764788734639e-05,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.22741346210241317,
|
|
"num_tokens": 17642119.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.624781227111816,
|
|
"epoch": 8.342071336484745,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.529748856285228e-05,
|
|
"loss": 4.9694,
|
|
"mean_token_accuracy": 0.22088632881641387,
|
|
"num_tokens": 17651612.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.6446921825408936,
|
|
"epoch": 8.346368715083798,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.511891248670217e-05,
|
|
"loss": 4.9086,
|
|
"mean_token_accuracy": 0.2222127616405487,
|
|
"num_tokens": 17660240.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.674179315567017,
|
|
"epoch": 8.350666093682854,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.494075103495245e-05,
|
|
"loss": 4.9356,
|
|
"mean_token_accuracy": 0.22276705801486968,
|
|
"num_tokens": 17669067.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.635128021240234,
|
|
"epoch": 8.354963472281907,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.476300459663643e-05,
|
|
"loss": 4.8644,
|
|
"mean_token_accuracy": 0.23134717643260955,
|
|
"num_tokens": 17677855.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.653323984146118,
|
|
"epoch": 8.359260850880963,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.458567355988123e-05,
|
|
"loss": 4.9605,
|
|
"mean_token_accuracy": 0.22503324449062348,
|
|
"num_tokens": 17686833.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.650402641296386,
|
|
"epoch": 8.363558229480017,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.440875831190704e-05,
|
|
"loss": 4.9322,
|
|
"mean_token_accuracy": 0.22219313383102418,
|
|
"num_tokens": 17695544.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.54823522567749,
|
|
"epoch": 8.367855608079072,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.423225923902608e-05,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.2386097714304924,
|
|
"num_tokens": 17704540.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.631325101852417,
|
|
"epoch": 8.372152986678126,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.405617672664175e-05,
|
|
"loss": 4.8472,
|
|
"mean_token_accuracy": 0.2270683839917183,
|
|
"num_tokens": 17713252.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.587792539596558,
|
|
"epoch": 8.376450365277181,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.388051115924781e-05,
|
|
"loss": 4.8059,
|
|
"mean_token_accuracy": 0.23830556273460388,
|
|
"num_tokens": 17722213.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.604350280761719,
|
|
"epoch": 8.380747743876235,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 8.370526292042774e-05,
|
|
"loss": 4.8983,
|
|
"mean_token_accuracy": 0.22606301605701445,
|
|
"num_tokens": 17730638.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.631150770187378,
|
|
"epoch": 8.38504512247529,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.353043239285357e-05,
|
|
"loss": 4.9211,
|
|
"mean_token_accuracy": 0.2233014479279518,
|
|
"num_tokens": 17739450.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.656581783294678,
|
|
"epoch": 8.389342501074344,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.335601995828534e-05,
|
|
"loss": 4.9345,
|
|
"mean_token_accuracy": 0.2134546771645546,
|
|
"num_tokens": 17748508.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.586218214035034,
|
|
"epoch": 8.3936398796734,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.318202599757008e-05,
|
|
"loss": 4.8966,
|
|
"mean_token_accuracy": 0.2175373390316963,
|
|
"num_tokens": 17758039.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.669969367980957,
|
|
"epoch": 8.397937258272453,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.30084508906411e-05,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.2146724358201027,
|
|
"num_tokens": 17768588.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.660363006591797,
|
|
"epoch": 8.402234636871508,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 8.283529501651698e-05,
|
|
"loss": 4.9531,
|
|
"mean_token_accuracy": 0.21941125988960267,
|
|
"num_tokens": 17777537.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.650928688049317,
|
|
"epoch": 8.406532015470564,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.266255875330095e-05,
|
|
"loss": 4.9258,
|
|
"mean_token_accuracy": 0.22859902679920197,
|
|
"num_tokens": 17787154.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.696631383895874,
|
|
"epoch": 8.410829394069617,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.249024247817998e-05,
|
|
"loss": 4.964,
|
|
"mean_token_accuracy": 0.2201056107878685,
|
|
"num_tokens": 17797058.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.690751075744629,
|
|
"epoch": 8.415126772668673,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.231834656742402e-05,
|
|
"loss": 4.9738,
|
|
"mean_token_accuracy": 0.21459610164165496,
|
|
"num_tokens": 17806389.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.635775852203369,
|
|
"epoch": 8.419424151267727,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.214687139638494e-05,
|
|
"loss": 4.9287,
|
|
"mean_token_accuracy": 0.22277025133371353,
|
|
"num_tokens": 17815878.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.713388204574585,
|
|
"epoch": 8.423721529866782,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.197581733949597e-05,
|
|
"loss": 5.0246,
|
|
"mean_token_accuracy": 0.20780490040779115,
|
|
"num_tokens": 17824883.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.611294937133789,
|
|
"epoch": 8.428018908465836,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.180518477027083e-05,
|
|
"loss": 4.8604,
|
|
"mean_token_accuracy": 0.23792800754308702,
|
|
"num_tokens": 17832691.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.652623176574707,
|
|
"epoch": 8.432316287064891,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.163497406130287e-05,
|
|
"loss": 4.9786,
|
|
"mean_token_accuracy": 0.2159315049648285,
|
|
"num_tokens": 17842270.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.652829837799072,
|
|
"epoch": 8.436613665663945,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.146518558426426e-05,
|
|
"loss": 4.9458,
|
|
"mean_token_accuracy": 0.2194216713309288,
|
|
"num_tokens": 17851268.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.652003288269043,
|
|
"epoch": 8.440911044263,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.129581970990507e-05,
|
|
"loss": 4.9721,
|
|
"mean_token_accuracy": 0.21328481137752534,
|
|
"num_tokens": 17861154.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.633577966690064,
|
|
"epoch": 8.445208422862054,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.11268768080528e-05,
|
|
"loss": 4.9442,
|
|
"mean_token_accuracy": 0.22499480694532395,
|
|
"num_tokens": 17869692.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.662738513946533,
|
|
"epoch": 8.44950580146111,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.09583572476111e-05,
|
|
"loss": 5.0043,
|
|
"mean_token_accuracy": 0.21218062490224837,
|
|
"num_tokens": 17879273.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.6502828121185305,
|
|
"epoch": 8.453803180060163,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.079026139655946e-05,
|
|
"loss": 4.98,
|
|
"mean_token_accuracy": 0.2203117623925209,
|
|
"num_tokens": 17887961.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.65530309677124,
|
|
"epoch": 8.458100558659218,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 8.062258962195192e-05,
|
|
"loss": 4.9831,
|
|
"mean_token_accuracy": 0.2141238272190094,
|
|
"num_tokens": 17897201.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.663492631912232,
|
|
"epoch": 8.462397937258272,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.04553422899167e-05,
|
|
"loss": 4.9449,
|
|
"mean_token_accuracy": 0.22099066525697708,
|
|
"num_tokens": 17906965.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.69427227973938,
|
|
"epoch": 8.466695315857327,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 8.028851976565508e-05,
|
|
"loss": 4.9886,
|
|
"mean_token_accuracy": 0.21574290841817856,
|
|
"num_tokens": 17915572.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.693426132202148,
|
|
"epoch": 8.470992694456381,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.01221224134408e-05,
|
|
"loss": 4.9655,
|
|
"mean_token_accuracy": 0.21988298743963242,
|
|
"num_tokens": 17924567.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.683645153045655,
|
|
"epoch": 8.475290073055437,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.995615059661907e-05,
|
|
"loss": 4.9619,
|
|
"mean_token_accuracy": 0.21657515466213226,
|
|
"num_tokens": 17932943.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.684688234329224,
|
|
"epoch": 8.47958745165449,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 7.979060467760616e-05,
|
|
"loss": 4.9169,
|
|
"mean_token_accuracy": 0.22091280221939086,
|
|
"num_tokens": 17941473.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.631937789916992,
|
|
"epoch": 8.483884830253546,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.962548501788811e-05,
|
|
"loss": 4.8799,
|
|
"mean_token_accuracy": 0.2327748417854309,
|
|
"num_tokens": 17950275.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.582111501693726,
|
|
"epoch": 8.4881822088526,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.946079197802028e-05,
|
|
"loss": 4.9069,
|
|
"mean_token_accuracy": 0.2211355298757553,
|
|
"num_tokens": 17959831.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.591459274291992,
|
|
"epoch": 8.492479587451655,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.929652591762636e-05,
|
|
"loss": 4.9403,
|
|
"mean_token_accuracy": 0.22110827565193175,
|
|
"num_tokens": 17969637.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.663167333602905,
|
|
"epoch": 8.496776966050708,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.913268719539785e-05,
|
|
"loss": 4.9377,
|
|
"mean_token_accuracy": 0.23038235306739807,
|
|
"num_tokens": 17978868.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.648486614227295,
|
|
"epoch": 8.501074344649764,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.896927616909304e-05,
|
|
"loss": 4.9347,
|
|
"mean_token_accuracy": 0.22413993775844573,
|
|
"num_tokens": 17986983.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.592524290084839,
|
|
"epoch": 8.505371723248818,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.880629319553623e-05,
|
|
"loss": 4.8825,
|
|
"mean_token_accuracy": 0.22498976290225983,
|
|
"num_tokens": 17996187.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.68932032585144,
|
|
"epoch": 8.509669101847873,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.864373863061717e-05,
|
|
"loss": 5.007,
|
|
"mean_token_accuracy": 0.21501995772123336,
|
|
"num_tokens": 18006243.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.6731819152832035,
|
|
"epoch": 8.513966480446927,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.848161282929006e-05,
|
|
"loss": 4.9405,
|
|
"mean_token_accuracy": 0.21863509714603424,
|
|
"num_tokens": 18015545.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.679911756515503,
|
|
"epoch": 8.518263859045982,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.831991614557274e-05,
|
|
"loss": 4.9523,
|
|
"mean_token_accuracy": 0.21662481874227524,
|
|
"num_tokens": 18024785.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.647496891021729,
|
|
"epoch": 8.522561237645036,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.815864893254619e-05,
|
|
"loss": 4.8864,
|
|
"mean_token_accuracy": 0.21894090473651887,
|
|
"num_tokens": 18033686.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.685502290725708,
|
|
"epoch": 8.526858616244091,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.799781154235361e-05,
|
|
"loss": 5.0639,
|
|
"mean_token_accuracy": 0.2153212159872055,
|
|
"num_tokens": 18044515.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.635985231399536,
|
|
"epoch": 8.531155994843147,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.783740432619954e-05,
|
|
"loss": 4.9703,
|
|
"mean_token_accuracy": 0.2184527799487114,
|
|
"num_tokens": 18054599.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.668966817855835,
|
|
"epoch": 8.5354533734422,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 7.767742763434922e-05,
|
|
"loss": 4.9394,
|
|
"mean_token_accuracy": 0.22269601225852967,
|
|
"num_tokens": 18063257.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.683325481414795,
|
|
"epoch": 8.539750752041256,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.75178818161277e-05,
|
|
"loss": 4.9796,
|
|
"mean_token_accuracy": 0.21864136010408403,
|
|
"num_tokens": 18071831.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.680855560302734,
|
|
"epoch": 8.54404813064031,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.735876721991945e-05,
|
|
"loss": 4.9596,
|
|
"mean_token_accuracy": 0.22199319452047347,
|
|
"num_tokens": 18080737.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.6189866065979,
|
|
"epoch": 8.548345509239365,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.720008419316708e-05,
|
|
"loss": 4.9663,
|
|
"mean_token_accuracy": 0.214399853348732,
|
|
"num_tokens": 18089254.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.61535496711731,
|
|
"epoch": 8.552642887838418,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.704183308237089e-05,
|
|
"loss": 4.9026,
|
|
"mean_token_accuracy": 0.2202185347676277,
|
|
"num_tokens": 18099027.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.675382900238037,
|
|
"epoch": 8.556940266437474,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.688401423308799e-05,
|
|
"loss": 4.9081,
|
|
"mean_token_accuracy": 0.21864477545022964,
|
|
"num_tokens": 18107326.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.617956447601318,
|
|
"epoch": 8.561237645036528,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.672662798993174e-05,
|
|
"loss": 4.8819,
|
|
"mean_token_accuracy": 0.22545797973871232,
|
|
"num_tokens": 18116838.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.552269744873047,
|
|
"epoch": 8.565535023635583,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.656967469657083e-05,
|
|
"loss": 4.8218,
|
|
"mean_token_accuracy": 0.23020118176937104,
|
|
"num_tokens": 18125559.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.621671295166015,
|
|
"epoch": 8.569832402234637,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 7.641315469572841e-05,
|
|
"loss": 4.9508,
|
|
"mean_token_accuracy": 0.21269513070583343,
|
|
"num_tokens": 18135010.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.661844110488891,
|
|
"epoch": 8.574129780833692,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.625706832918172e-05,
|
|
"loss": 4.9794,
|
|
"mean_token_accuracy": 0.2107702150940895,
|
|
"num_tokens": 18144377.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.640846252441406,
|
|
"epoch": 8.578427159432746,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.610141593776091e-05,
|
|
"loss": 4.8924,
|
|
"mean_token_accuracy": 0.22317289113998412,
|
|
"num_tokens": 18152962.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.658638000488281,
|
|
"epoch": 8.582724538031801,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.59461978613486e-05,
|
|
"loss": 4.961,
|
|
"mean_token_accuracy": 0.21346425265073776,
|
|
"num_tokens": 18162692.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.699154853820801,
|
|
"epoch": 8.587021916630855,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 7.579141443887901e-05,
|
|
"loss": 5.0204,
|
|
"mean_token_accuracy": 0.21661454439163208,
|
|
"num_tokens": 18172154.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.7034563541412355,
|
|
"epoch": 8.59131929522991,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.563706600833737e-05,
|
|
"loss": 4.9581,
|
|
"mean_token_accuracy": 0.21558541655540467,
|
|
"num_tokens": 18180373.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 8.59131929522991,
|
|
"eval_entropy": 5.5324751917306365,
|
|
"eval_loss": 5.940296649932861,
|
|
"eval_mean_token_accuracy": 0.1781518642228466,
|
|
"eval_num_tokens": 18180373.0,
|
|
"eval_runtime": 2.6403,
|
|
"eval_samples_per_second": 1344.184,
|
|
"eval_steps_per_second": 168.165,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.7329833030700685,
|
|
"epoch": 8.595616673828964,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.548315290675886e-05,
|
|
"loss": 4.9755,
|
|
"mean_token_accuracy": 0.21385726928710938,
|
|
"num_tokens": 18189976.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.670011329650879,
|
|
"epoch": 8.59991405242802,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 7.532967547022825e-05,
|
|
"loss": 4.942,
|
|
"mean_token_accuracy": 0.22368471324443817,
|
|
"num_tokens": 18198273.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.645975732803345,
|
|
"epoch": 8.604211431027073,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.517663403387874e-05,
|
|
"loss": 4.9102,
|
|
"mean_token_accuracy": 0.2267240732908249,
|
|
"num_tokens": 18206119.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.627954435348511,
|
|
"epoch": 8.608508809626128,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.502402893189191e-05,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.21353731006383897,
|
|
"num_tokens": 18215124.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.659151268005371,
|
|
"epoch": 8.612806188225182,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.487186049749618e-05,
|
|
"loss": 4.9897,
|
|
"mean_token_accuracy": 0.2172723338007927,
|
|
"num_tokens": 18224956.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.65262279510498,
|
|
"epoch": 8.617103566824238,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.472012906296658e-05,
|
|
"loss": 4.9059,
|
|
"mean_token_accuracy": 0.2237917348742485,
|
|
"num_tokens": 18232472.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.614779663085938,
|
|
"epoch": 8.621400945423291,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.45688349596239e-05,
|
|
"loss": 4.9663,
|
|
"mean_token_accuracy": 0.21777795106172562,
|
|
"num_tokens": 18242008.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.611991453170776,
|
|
"epoch": 8.625698324022347,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.441797851783402e-05,
|
|
"loss": 4.8554,
|
|
"mean_token_accuracy": 0.23138580918312074,
|
|
"num_tokens": 18252091.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.61883282661438,
|
|
"epoch": 8.6299957026214,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.426756006700716e-05,
|
|
"loss": 4.8845,
|
|
"mean_token_accuracy": 0.22362819761037828,
|
|
"num_tokens": 18260718.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.654276132583618,
|
|
"epoch": 8.634293081220456,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.411757993559702e-05,
|
|
"loss": 4.9095,
|
|
"mean_token_accuracy": 0.22516279369592668,
|
|
"num_tokens": 18269857.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.64529595375061,
|
|
"epoch": 8.63859045981951,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.396803845110032e-05,
|
|
"loss": 4.9592,
|
|
"mean_token_accuracy": 0.218281888961792,
|
|
"num_tokens": 18280166.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.646187591552734,
|
|
"epoch": 8.642887838418565,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.381893594005585e-05,
|
|
"loss": 4.9086,
|
|
"mean_token_accuracy": 0.22456822097301482,
|
|
"num_tokens": 18288982.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.567366075515747,
|
|
"epoch": 8.647185217017618,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.367027272804387e-05,
|
|
"loss": 4.8264,
|
|
"mean_token_accuracy": 0.23126058727502824,
|
|
"num_tokens": 18298332.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.7077497959136965,
|
|
"epoch": 8.651482595616674,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.352204913968546e-05,
|
|
"loss": 5.0316,
|
|
"mean_token_accuracy": 0.2043105497956276,
|
|
"num_tokens": 18307835.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.591073131561279,
|
|
"epoch": 8.65577997421573,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.337426549864175e-05,
|
|
"loss": 4.9514,
|
|
"mean_token_accuracy": 0.218123097717762,
|
|
"num_tokens": 18316889.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.652389812469482,
|
|
"epoch": 8.660077352814783,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.322692212761306e-05,
|
|
"loss": 4.9662,
|
|
"mean_token_accuracy": 0.22802242934703826,
|
|
"num_tokens": 18326826.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.628634405136109,
|
|
"epoch": 8.664374731413837,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.308001934833844e-05,
|
|
"loss": 4.8484,
|
|
"mean_token_accuracy": 0.22803269773721696,
|
|
"num_tokens": 18334638.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.713373756408691,
|
|
"epoch": 8.668672110012892,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 7.29335574815948e-05,
|
|
"loss": 4.9664,
|
|
"mean_token_accuracy": 0.21574983447790147,
|
|
"num_tokens": 18343831.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.681832885742187,
|
|
"epoch": 8.672969488611947,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.278753684719636e-05,
|
|
"loss": 4.9872,
|
|
"mean_token_accuracy": 0.22094194889068602,
|
|
"num_tokens": 18352734.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.604401636123657,
|
|
"epoch": 8.677266867211001,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.264195776399386e-05,
|
|
"loss": 4.8892,
|
|
"mean_token_accuracy": 0.2287626013159752,
|
|
"num_tokens": 18361788.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.692612409591675,
|
|
"epoch": 8.681564245810057,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.249682054987381e-05,
|
|
"loss": 4.975,
|
|
"mean_token_accuracy": 0.22251248359680176,
|
|
"num_tokens": 18370319.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.6745744228363035,
|
|
"epoch": 8.68586162440911,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.23521255217578e-05,
|
|
"loss": 5.0155,
|
|
"mean_token_accuracy": 0.21344987601041793,
|
|
"num_tokens": 18378958.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.6371691703796385,
|
|
"epoch": 8.690159003008166,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.220787299560205e-05,
|
|
"loss": 4.9289,
|
|
"mean_token_accuracy": 0.22289830893278123,
|
|
"num_tokens": 18387510.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.596817541122436,
|
|
"epoch": 8.69445638160722,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 7.20640632863963e-05,
|
|
"loss": 4.9638,
|
|
"mean_token_accuracy": 0.21788981407880784,
|
|
"num_tokens": 18396484.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.5960277080535885,
|
|
"epoch": 8.698753760206275,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.192069670816359e-05,
|
|
"loss": 4.8113,
|
|
"mean_token_accuracy": 0.23225875347852706,
|
|
"num_tokens": 18405413.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.695937538146973,
|
|
"epoch": 8.703051138805328,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.177777357395912e-05,
|
|
"loss": 4.9726,
|
|
"mean_token_accuracy": 0.21074809283018112,
|
|
"num_tokens": 18414352.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.67652587890625,
|
|
"epoch": 8.707348517404384,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.163529419587002e-05,
|
|
"loss": 5.0142,
|
|
"mean_token_accuracy": 0.21143629550933837,
|
|
"num_tokens": 18422731.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.687342834472656,
|
|
"epoch": 8.711645896003438,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.149325888501418e-05,
|
|
"loss": 4.9054,
|
|
"mean_token_accuracy": 0.2306586056947708,
|
|
"num_tokens": 18432023.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.673311519622803,
|
|
"epoch": 8.715943274602493,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.135166795153992e-05,
|
|
"loss": 4.9912,
|
|
"mean_token_accuracy": 0.21262432634830475,
|
|
"num_tokens": 18440483.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.630145692825318,
|
|
"epoch": 8.720240653201547,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.121052170462541e-05,
|
|
"loss": 4.9045,
|
|
"mean_token_accuracy": 0.2280057206749916,
|
|
"num_tokens": 18450830.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.621188831329346,
|
|
"epoch": 8.724538031800602,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.106982045247754e-05,
|
|
"loss": 4.8971,
|
|
"mean_token_accuracy": 0.22428022921085358,
|
|
"num_tokens": 18460625.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.6501161575317385,
|
|
"epoch": 8.728835410399656,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.092956450233162e-05,
|
|
"loss": 4.959,
|
|
"mean_token_accuracy": 0.21805860698223115,
|
|
"num_tokens": 18470254.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.615547227859497,
|
|
"epoch": 8.733132788998711,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.078975416045055e-05,
|
|
"loss": 4.9392,
|
|
"mean_token_accuracy": 0.21594141721725463,
|
|
"num_tokens": 18479762.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.645567607879639,
|
|
"epoch": 8.737430167597765,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.065038973212424e-05,
|
|
"loss": 4.9449,
|
|
"mean_token_accuracy": 0.2187328338623047,
|
|
"num_tokens": 18489592.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.594530010223389,
|
|
"epoch": 8.74172754619682,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.051147152166896e-05,
|
|
"loss": 4.8824,
|
|
"mean_token_accuracy": 0.21929869800806046,
|
|
"num_tokens": 18498553.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.6663798809051515,
|
|
"epoch": 8.746024924795874,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 7.037299983242652e-05,
|
|
"loss": 4.9938,
|
|
"mean_token_accuracy": 0.21613374203443528,
|
|
"num_tokens": 18506817.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.595138931274414,
|
|
"epoch": 8.75032230339493,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.023497496676371e-05,
|
|
"loss": 4.8531,
|
|
"mean_token_accuracy": 0.22364323288202287,
|
|
"num_tokens": 18515478.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.670313739776612,
|
|
"epoch": 8.754619681993983,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 7.009739722607173e-05,
|
|
"loss": 4.9942,
|
|
"mean_token_accuracy": 0.2162443593144417,
|
|
"num_tokens": 18524553.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.666472148895264,
|
|
"epoch": 8.758917060593038,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.996026691076531e-05,
|
|
"loss": 4.983,
|
|
"mean_token_accuracy": 0.22348050624132157,
|
|
"num_tokens": 18533626.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.671681356430054,
|
|
"epoch": 8.763214439192092,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.982358432028234e-05,
|
|
"loss": 4.9486,
|
|
"mean_token_accuracy": 0.22210537642240524,
|
|
"num_tokens": 18542509.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.673493480682373,
|
|
"epoch": 8.767511817791148,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.968734975308283e-05,
|
|
"loss": 4.9835,
|
|
"mean_token_accuracy": 0.21366600692272186,
|
|
"num_tokens": 18552513.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.625103616714478,
|
|
"epoch": 8.771809196390201,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.955156350664876e-05,
|
|
"loss": 4.9137,
|
|
"mean_token_accuracy": 0.21497104167938233,
|
|
"num_tokens": 18561148.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.642923498153687,
|
|
"epoch": 8.776106574989257,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.941622587748298e-05,
|
|
"loss": 4.9828,
|
|
"mean_token_accuracy": 0.22014665454626084,
|
|
"num_tokens": 18570198.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.587640714645386,
|
|
"epoch": 8.780403953588312,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.928133716110863e-05,
|
|
"loss": 4.8689,
|
|
"mean_token_accuracy": 0.22607097774744034,
|
|
"num_tokens": 18580049.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.59015154838562,
|
|
"epoch": 8.784701332187366,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.914689765206899e-05,
|
|
"loss": 4.9247,
|
|
"mean_token_accuracy": 0.22805389761924744,
|
|
"num_tokens": 18590128.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.681815481185913,
|
|
"epoch": 8.78899871078642,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.901290764392609e-05,
|
|
"loss": 5.0546,
|
|
"mean_token_accuracy": 0.20661425590515137,
|
|
"num_tokens": 18599485.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.735357236862183,
|
|
"epoch": 8.793296089385475,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.887936742926058e-05,
|
|
"loss": 5.0036,
|
|
"mean_token_accuracy": 0.21456173062324524,
|
|
"num_tokens": 18608833.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.609775876998901,
|
|
"epoch": 8.79759346798453,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.874627729967086e-05,
|
|
"loss": 4.9144,
|
|
"mean_token_accuracy": 0.23087783008813859,
|
|
"num_tokens": 18617315.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.592963933944702,
|
|
"epoch": 8.801890846583584,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.861363754577262e-05,
|
|
"loss": 4.9753,
|
|
"mean_token_accuracy": 0.22231801450252534,
|
|
"num_tokens": 18626492.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.657146215438843,
|
|
"epoch": 8.80618822518264,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.848144845719808e-05,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.22680884450674058,
|
|
"num_tokens": 18634986.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.632013034820557,
|
|
"epoch": 8.810485603781693,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.834971032259537e-05,
|
|
"loss": 4.9724,
|
|
"mean_token_accuracy": 0.21331976652145385,
|
|
"num_tokens": 18644767.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.685769319534302,
|
|
"epoch": 8.814782982380748,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.821842342962786e-05,
|
|
"loss": 4.8918,
|
|
"mean_token_accuracy": 0.22922518104314804,
|
|
"num_tokens": 18652278.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.668872928619384,
|
|
"epoch": 8.819080360979802,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.808758806497375e-05,
|
|
"loss": 4.9225,
|
|
"mean_token_accuracy": 0.2232070282101631,
|
|
"num_tokens": 18660996.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.636686897277832,
|
|
"epoch": 8.823377739578858,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.795720451432509e-05,
|
|
"loss": 4.9765,
|
|
"mean_token_accuracy": 0.2159225896000862,
|
|
"num_tokens": 18670406.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.615257930755615,
|
|
"epoch": 8.827675118177911,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.782727306238749e-05,
|
|
"loss": 4.8517,
|
|
"mean_token_accuracy": 0.23174299001693727,
|
|
"num_tokens": 18678663.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.662962484359741,
|
|
"epoch": 8.831972496776967,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.769779399287928e-05,
|
|
"loss": 4.8894,
|
|
"mean_token_accuracy": 0.22123230695724488,
|
|
"num_tokens": 18688005.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.575050306320191,
|
|
"epoch": 8.83626987537602,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.756876758853104e-05,
|
|
"loss": 4.8916,
|
|
"mean_token_accuracy": 0.2288811132311821,
|
|
"num_tokens": 18696900.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.714194440841675,
|
|
"epoch": 8.840567253975076,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 6.744019413108486e-05,
|
|
"loss": 5.0764,
|
|
"mean_token_accuracy": 0.21059815883636473,
|
|
"num_tokens": 18706865.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.655612373352051,
|
|
"epoch": 8.84486463257413,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.731207390129366e-05,
|
|
"loss": 4.9713,
|
|
"mean_token_accuracy": 0.2176967516541481,
|
|
"num_tokens": 18716554.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.589277076721191,
|
|
"epoch": 8.849162011173185,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.7184407178921e-05,
|
|
"loss": 4.9303,
|
|
"mean_token_accuracy": 0.2208853080868721,
|
|
"num_tokens": 18725857.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.665862703323365,
|
|
"epoch": 8.853459389772238,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.70571942427399e-05,
|
|
"loss": 4.965,
|
|
"mean_token_accuracy": 0.21218861341476442,
|
|
"num_tokens": 18735137.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.678767108917237,
|
|
"epoch": 8.857756768371294,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.693043537053254e-05,
|
|
"loss": 4.9855,
|
|
"mean_token_accuracy": 0.2140743687748909,
|
|
"num_tokens": 18744843.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.649939393997192,
|
|
"epoch": 8.862054146970348,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.68041308390896e-05,
|
|
"loss": 4.8776,
|
|
"mean_token_accuracy": 0.22999678552150726,
|
|
"num_tokens": 18753375.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.670957946777344,
|
|
"epoch": 8.866351525569403,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.667828092420977e-05,
|
|
"loss": 4.9409,
|
|
"mean_token_accuracy": 0.22305206209421158,
|
|
"num_tokens": 18762034.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.697635269165039,
|
|
"epoch": 8.870648904168457,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.655288590069889e-05,
|
|
"loss": 4.9881,
|
|
"mean_token_accuracy": 0.2155222401022911,
|
|
"num_tokens": 18771254.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.682199716567993,
|
|
"epoch": 8.874946282767512,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.642794604236965e-05,
|
|
"loss": 5.0381,
|
|
"mean_token_accuracy": 0.20920461267232895,
|
|
"num_tokens": 18780160.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.67813720703125,
|
|
"epoch": 8.879243661366566,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.630346162204069e-05,
|
|
"loss": 5.0109,
|
|
"mean_token_accuracy": 0.21448171585798265,
|
|
"num_tokens": 18790270.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.634468746185303,
|
|
"epoch": 8.883541039965621,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.617943291153631e-05,
|
|
"loss": 4.971,
|
|
"mean_token_accuracy": 0.22084443867206574,
|
|
"num_tokens": 18799300.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.664403057098388,
|
|
"epoch": 8.887838418564675,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.605586018168558e-05,
|
|
"loss": 4.9354,
|
|
"mean_token_accuracy": 0.21384454369544983,
|
|
"num_tokens": 18808189.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.571662473678589,
|
|
"epoch": 8.89213579716373,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.593274370232191e-05,
|
|
"loss": 4.8763,
|
|
"mean_token_accuracy": 0.2294844552874565,
|
|
"num_tokens": 18818287.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.679098272323609,
|
|
"epoch": 8.896433175762784,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.581008374228255e-05,
|
|
"loss": 4.966,
|
|
"mean_token_accuracy": 0.2185201585292816,
|
|
"num_tokens": 18827528.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.631080389022827,
|
|
"epoch": 8.90073055436184,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 6.568788056940785e-05,
|
|
"loss": 4.8856,
|
|
"mean_token_accuracy": 0.22746642380952836,
|
|
"num_tokens": 18836274.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.616071701049805,
|
|
"epoch": 8.905027932960895,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.556613445054063e-05,
|
|
"loss": 4.8883,
|
|
"mean_token_accuracy": 0.2287377655506134,
|
|
"num_tokens": 18845383.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.579406309127807,
|
|
"epoch": 8.909325311559948,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.544484565152577e-05,
|
|
"loss": 4.8135,
|
|
"mean_token_accuracy": 0.23297637403011323,
|
|
"num_tokens": 18853415.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.560865116119385,
|
|
"epoch": 8.913622690159002,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.532401443720951e-05,
|
|
"loss": 4.9281,
|
|
"mean_token_accuracy": 0.21631055027246476,
|
|
"num_tokens": 18862537.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.55099458694458,
|
|
"epoch": 8.917920068758058,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.520364107143898e-05,
|
|
"loss": 4.8785,
|
|
"mean_token_accuracy": 0.22973405122756957,
|
|
"num_tokens": 18871654.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.5747205257415775,
|
|
"epoch": 8.922217447357113,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.50837258170615e-05,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.23019151985645295,
|
|
"num_tokens": 18879877.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.622156620025635,
|
|
"epoch": 8.926514825956167,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.496426893592397e-05,
|
|
"loss": 4.9419,
|
|
"mean_token_accuracy": 0.21682817935943605,
|
|
"num_tokens": 18889537.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.677370882034301,
|
|
"epoch": 8.930812204555222,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.484527068887258e-05,
|
|
"loss": 4.9907,
|
|
"mean_token_accuracy": 0.21561216861009597,
|
|
"num_tokens": 18898360.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.65316481590271,
|
|
"epoch": 8.935109583154276,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.472673133575181e-05,
|
|
"loss": 4.9692,
|
|
"mean_token_accuracy": 0.21621489375829697,
|
|
"num_tokens": 18907519.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.594114637374878,
|
|
"epoch": 8.939406961753331,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.460865113540437e-05,
|
|
"loss": 4.8859,
|
|
"mean_token_accuracy": 0.22777420282363892,
|
|
"num_tokens": 18916278.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.588778638839722,
|
|
"epoch": 8.943704340352385,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 6.449103034567011e-05,
|
|
"loss": 4.8679,
|
|
"mean_token_accuracy": 0.22822130620479583,
|
|
"num_tokens": 18925567.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.661559534072876,
|
|
"epoch": 8.94800171895144,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.437386922338591e-05,
|
|
"loss": 4.9536,
|
|
"mean_token_accuracy": 0.22029758244752884,
|
|
"num_tokens": 18934487.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.704027080535889,
|
|
"epoch": 8.952299097550494,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.425716802438479e-05,
|
|
"loss": 5.0275,
|
|
"mean_token_accuracy": 0.20596762001514435,
|
|
"num_tokens": 18942787.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.593880033493042,
|
|
"epoch": 8.95659647614955,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.414092700349548e-05,
|
|
"loss": 4.9223,
|
|
"mean_token_accuracy": 0.220767804980278,
|
|
"num_tokens": 18951610.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.643803501129151,
|
|
"epoch": 8.960893854748603,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.402514641454192e-05,
|
|
"loss": 5.0342,
|
|
"mean_token_accuracy": 0.22101122438907622,
|
|
"num_tokens": 18961048.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.626805830001831,
|
|
"epoch": 8.965191233347658,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.390982651034274e-05,
|
|
"loss": 4.9393,
|
|
"mean_token_accuracy": 0.22341009825468064,
|
|
"num_tokens": 18969744.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.703090620040894,
|
|
"epoch": 8.969488611946712,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.379496754271044e-05,
|
|
"loss": 4.997,
|
|
"mean_token_accuracy": 0.2205844923853874,
|
|
"num_tokens": 18979733.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.729594564437866,
|
|
"epoch": 8.973785990545768,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.368056976245107e-05,
|
|
"loss": 5.0159,
|
|
"mean_token_accuracy": 0.21660550087690353,
|
|
"num_tokens": 18989154.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.666148328781128,
|
|
"epoch": 8.978083369144821,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.356663341936368e-05,
|
|
"loss": 5.0295,
|
|
"mean_token_accuracy": 0.2158375933766365,
|
|
"num_tokens": 18998553.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.6620241641998295,
|
|
"epoch": 8.982380747743877,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 6.345315876223977e-05,
|
|
"loss": 4.9033,
|
|
"mean_token_accuracy": 0.22086922079324722,
|
|
"num_tokens": 19006482.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.5746057510375975,
|
|
"epoch": 8.98667812634293,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.334014603886256e-05,
|
|
"loss": 4.838,
|
|
"mean_token_accuracy": 0.22812474220991136,
|
|
"num_tokens": 19014850.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.683941793441773,
|
|
"epoch": 8.990975504941986,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.322759549600665e-05,
|
|
"loss": 4.9742,
|
|
"mean_token_accuracy": 0.21915542483329772,
|
|
"num_tokens": 19024074.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.629464721679687,
|
|
"epoch": 8.99527288354104,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.311550737943753e-05,
|
|
"loss": 4.9045,
|
|
"mean_token_accuracy": 0.22415376156568528,
|
|
"num_tokens": 19033821.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.563567686080932,
|
|
"epoch": 8.999570262140095,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.300388193391075e-05,
|
|
"loss": 4.8629,
|
|
"mean_token_accuracy": 0.22649890929460526,
|
|
"num_tokens": 19042275.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.673954274919298,
|
|
"epoch": 9.003437902879243,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.289271940317174e-05,
|
|
"loss": 4.9573,
|
|
"mean_token_accuracy": 0.21372253530555302,
|
|
"num_tokens": 19050103.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.609748220443725,
|
|
"epoch": 9.007735281478299,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.278202002995497e-05,
|
|
"loss": 4.8947,
|
|
"mean_token_accuracy": 0.21994215548038482,
|
|
"num_tokens": 19059843.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.650927782058716,
|
|
"epoch": 9.012032660077352,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.267178405598371e-05,
|
|
"loss": 4.9354,
|
|
"mean_token_accuracy": 0.22398955821990968,
|
|
"num_tokens": 19069030.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.6360210418701175,
|
|
"epoch": 9.016330038676408,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.256201172196921e-05,
|
|
"loss": 4.8752,
|
|
"mean_token_accuracy": 0.21827268302440644,
|
|
"num_tokens": 19077754.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.655957841873169,
|
|
"epoch": 9.020627417275461,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.245270326761034e-05,
|
|
"loss": 4.8311,
|
|
"mean_token_accuracy": 0.2320619285106659,
|
|
"num_tokens": 19087425.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 9.020627417275461,
|
|
"eval_entropy": 5.518933880973506,
|
|
"eval_loss": 5.939923286437988,
|
|
"eval_mean_token_accuracy": 0.17867249829342235,
|
|
"eval_num_tokens": 19087425.0,
|
|
"eval_runtime": 2.8409,
|
|
"eval_samples_per_second": 1249.262,
|
|
"eval_steps_per_second": 156.29,
|
|
"step": 10500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11630,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2.893350462336e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|