tinystories-diff-ar / refill_results.json
Asilarkness's picture
Super-squash branch 'main' using huggingface_hub
a6c51a2
Raw History Blame Contribute Delete
6.21 kB
[
{
"scheme": "AR (ceiling)",
"calls": 128,
"sec": 0.3469,
"tok_s": 23617,
"nll": 2.8144,
"dup": 0.0177,
"d2": 0.5155
},
{
"scheme": "DIFF K=2",
"calls": 2,
"sec": 0.0159,
"tok_s": 515364,
"nll": 8.422,
"dup": 0.0495,
"d2": 0.5907
},
{
"scheme": "DIFF K=4",
"calls": 4,
"sec": 0.0319,
"tok_s": 256512,
"nll": 5.8584,
"dup": 0.0586,
"d2": 0.4783
},
{
"scheme": "DIFF K=6",
"calls": 6,
"sec": 0.0472,
"tok_s": 173597,
"nll": 4.8534,
"dup": 0.0607,
"d2": 0.4547
},
{
"scheme": "DIFF K=8",
"calls": 8,
"sec": 0.0625,
"tok_s": 131150,
"nll": 4.3593,
"dup": 0.0535,
"d2": 0.4401
},
{
"scheme": "DIFF K=10",
"calls": 10,
"sec": 0.0783,
"tok_s": 104655,
"nll": 4.1867,
"dup": 0.0544,
"d2": 0.4279
},
{
"scheme": "DIFF K=12",
"calls": 12,
"sec": 0.094,
"tok_s": 87121,
"nll": 3.8919,
"dup": 0.0528,
"d2": 0.4172
},
{
"scheme": "DIFF K=16",
"calls": 16,
"sec": 0.1246,
"tok_s": 65766,
"nll": 3.5531,
"dup": 0.046,
"d2": 0.4199
},
{
"scheme": "DIFF K=20",
"calls": 20,
"sec": 0.1564,
"tok_s": 52366,
"nll": 3.253,
"dup": 0.0433,
"d2": 0.4119
},
{
"scheme": "DIFF K=24",
"calls": 24,
"sec": 0.1866,
"tok_s": 43903,
"nll": 2.9899,
"dup": 0.0392,
"d2": 0.3936
},
{
"scheme": "K=4 p=0.10 R=1",
"calls": 5,
"sec": 0.037,
"tok_s": 221113,
"nll": 4.9766,
"dup": 0.0582,
"d2": 0.4325
},
{
"scheme": "K=4 p=0.10 R=2",
"calls": 6,
"sec": 0.0428,
"tok_s": 191548,
"nll": 4.7899,
"dup": 0.0594,
"d2": 0.4222
},
{
"scheme": "K=4 p=0.10 R=4",
"calls": 8,
"sec": 0.0543,
"tok_s": 150894,
"nll": 4.3275,
"dup": 0.0479,
"d2": 0.4278
},
{
"scheme": "K=4 p=0.20 R=1",
"calls": 5,
"sec": 0.0373,
"tok_s": 219508,
"nll": 4.976,
"dup": 0.0571,
"d2": 0.4402
},
{
"scheme": "K=4 p=0.20 R=2",
"calls": 6,
"sec": 0.0433,
"tok_s": 189256,
"nll": 4.3944,
"dup": 0.0486,
"d2": 0.4386
},
{
"scheme": "K=4 p=0.20 R=4",
"calls": 8,
"sec": 0.0549,
"tok_s": 149345,
"nll": 3.6246,
"dup": 0.0422,
"d2": 0.4466
},
{
"scheme": "K=4 p=0.30 R=1",
"calls": 5,
"sec": 0.0374,
"tok_s": 219003,
"nll": 4.7735,
"dup": 0.057,
"d2": 0.4427
},
{
"scheme": "K=4 p=0.30 R=2",
"calls": 6,
"sec": 0.0435,
"tok_s": 188193,
"nll": 4.1703,
"dup": 0.042,
"d2": 0.4502
},
{
"scheme": "K=4 p=0.30 R=4",
"calls": 8,
"sec": 0.0559,
"tok_s": 146419,
"nll": 3.5979,
"dup": 0.0402,
"d2": 0.4345
},
{
"scheme": "K=4 p=0.50 R=1",
"calls": 5,
"sec": 0.0379,
"tok_s": 216238,
"nll": 5.12,
"dup": 0.046,
"d2": 0.4922
},
{
"scheme": "K=4 p=0.50 R=2",
"calls": 6,
"sec": 0.0446,
"tok_s": 183568,
"nll": 4.7294,
"dup": 0.0399,
"d2": 0.5059
},
{
"scheme": "K=4 p=0.50 R=4",
"calls": 8,
"sec": 0.0577,
"tok_s": 142003,
"nll": 4.9303,
"dup": 0.044,
"d2": 0.5007
},
{
"scheme": "K=8 p=0.10 R=1",
"calls": 9,
"sec": 0.0682,
"tok_s": 120178,
"nll": 3.9082,
"dup": 0.0519,
"d2": 0.4197
},
{
"scheme": "K=8 p=0.10 R=2",
"calls": 10,
"sec": 0.074,
"tok_s": 110647,
"nll": 3.7144,
"dup": 0.0493,
"d2": 0.4158
},
{
"scheme": "K=8 p=0.10 R=4",
"calls": 12,
"sec": 0.0853,
"tok_s": 96021,
"nll": 3.3582,
"dup": 0.0415,
"d2": 0.4229
},
{
"scheme": "K=8 p=0.20 R=1",
"calls": 9,
"sec": 0.0686,
"tok_s": 119364,
"nll": 3.8277,
"dup": 0.049,
"d2": 0.428
},
{
"scheme": "K=8 p=0.20 R=2",
"calls": 10,
"sec": 0.0743,
"tok_s": 110238,
"nll": 3.4856,
"dup": 0.0395,
"d2": 0.4192
},
{
"scheme": "K=8 p=0.20 R=4",
"calls": 12,
"sec": 0.0857,
"tok_s": 95562,
"nll": 3.0254,
"dup": 0.0352,
"d2": 0.4211
},
{
"scheme": "K=8 p=0.30 R=1",
"calls": 9,
"sec": 0.0686,
"tok_s": 119427,
"nll": 4.0013,
"dup": 0.0464,
"d2": 0.4438
},
{
"scheme": "K=8 p=0.30 R=2",
"calls": 10,
"sec": 0.0744,
"tok_s": 110039,
"nll": 3.4876,
"dup": 0.0378,
"d2": 0.4374
},
{
"scheme": "K=8 p=0.30 R=4",
"calls": 12,
"sec": 0.0866,
"tok_s": 94619,
"nll": 3.3252,
"dup": 0.0338,
"d2": 0.4433
},
{
"scheme": "K=8 p=0.50 R=1",
"calls": 9,
"sec": 0.0693,
"tok_s": 118288,
"nll": 4.0792,
"dup": 0.0436,
"d2": 0.4488
},
{
"scheme": "K=8 p=0.50 R=2",
"calls": 10,
"sec": 0.076,
"tok_s": 107800,
"nll": 4.3721,
"dup": 0.0381,
"d2": 0.4774
},
{
"scheme": "K=8 p=0.50 R=4",
"calls": 12,
"sec": 0.0899,
"tok_s": 91114,
"nll": 4.6417,
"dup": 0.0426,
"d2": 0.4953
},
{
"scheme": "K=12 p=0.10 R=1",
"calls": 13,
"sec": 0.0993,
"tok_s": 82532,
"nll": 3.6624,
"dup": 0.0503,
"d2": 0.417
},
{
"scheme": "K=12 p=0.10 R=2",
"calls": 14,
"sec": 0.1058,
"tok_s": 77422,
"nll": 3.3433,
"dup": 0.0391,
"d2": 0.4181
},
{
"scheme": "K=12 p=0.10 R=4",
"calls": 16,
"sec": 0.1154,
"tok_s": 70984,
"nll": 3.0165,
"dup": 0.0351,
"d2": 0.4262
},
{
"scheme": "K=12 p=0.20 R=1",
"calls": 13,
"sec": 0.0993,
"tok_s": 82501,
"nll": 3.3954,
"dup": 0.0423,
"d2": 0.4222
},
{
"scheme": "K=12 p=0.20 R=2",
"calls": 14,
"sec": 0.1051,
"tok_s": 77928,
"nll": 3.2046,
"dup": 0.0383,
"d2": 0.4235
},
{
"scheme": "K=12 p=0.20 R=4",
"calls": 16,
"sec": 0.1173,
"tok_s": 69812,
"nll": 2.7801,
"dup": 0.0315,
"d2": 0.4131
},
{
"scheme": "K=12 p=0.30 R=1",
"calls": 13,
"sec": 0.1001,
"tok_s": 81803,
"nll": 3.4976,
"dup": 0.0386,
"d2": 0.4461
},
{
"scheme": "K=12 p=0.30 R=2",
"calls": 14,
"sec": 0.1054,
"tok_s": 77752,
"nll": 3.1743,
"dup": 0.0372,
"d2": 0.4304
},
{
"scheme": "K=12 p=0.30 R=4",
"calls": 16,
"sec": 0.1174,
"tok_s": 69781,
"nll": 3.1914,
"dup": 0.0283,
"d2": 0.4381
},
{
"scheme": "K=12 p=0.50 R=1",
"calls": 13,
"sec": 0.0999,
"tok_s": 82015,
"nll": 3.8986,
"dup": 0.0328,
"d2": 0.4703
},
{
"scheme": "K=12 p=0.50 R=2",
"calls": 14,
"sec": 0.1069,
"tok_s": 76610,
"nll": 4.1343,
"dup": 0.0346,
"d2": 0.4808
},
{
"scheme": "K=12 p=0.50 R=4",
"calls": 16,
"sec": 0.1196,
"tok_s": 68478,
"nll": 4.5074,
"dup": 0.0343,
"d2": 0.5069
}
]