List your company

Language Modeling on C4 (Entropy, Perplexity)

1EntropyBBQ
0.88281.67392.4653.2561Mar 2, 2026

Evaluation results

MethodLinks
BBQParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=12026.03149.22
QuESTParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=12026.03167.78
BBQParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=12026.03144.58
QuESTParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=12026.03172.54
BBQParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=12026.03138.27
QuESTParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=12026.03152.37
LSQParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=22026.031.6378.19
LSQParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=22026.031.6936.58
LSQParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=22026.031.8131.42
QuESTParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=22026.031.9235.58
QuESTParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=22026.031.9331.32
QuESTParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=22026.031.9325.46
QuESTParams=300M, Tokens=20B, Full Perplexity=15.43, Bits=22026.031.9321.53
BBQParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=22026.031.9731.34
BBQParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=22026.031.9827.34
BBQParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=22026.031.9823.08
BBQParams=300M, Tokens=20B, Full Perplexity=15.43, Bits=22026.031.9819.75
LSQParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=32026.032.5140.9
LSQParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=42026.032.731,778
LSQParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=32026.032.7430.27
LSQParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=32026.032.7426.28
QuESTParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=32026.032.7829.04
QuESTParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=32026.032.7825.21
QuESTParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=32026.032.7820.82
QuESTParams=300M, Tokens=20B, Full Perplexity=15.43, Bits=32026.032.7817.67
BBQParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=32026.032.9626.55
BBQParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=32026.032.9623.22
BBQParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=32026.032.9619.74
BBQParams=300M, Tokens=20B, Full Perplexity=15.43, Bits=32026.032.9616.9
LSQParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=42026.033.5927.46
LSQParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=42026.033.623.77
QuESTParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=42026.033.6126.37
QuESTParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=42026.033.6122.98
QuESTParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=42026.033.6119.06
QuESTParams=300M, Tokens=20B, Full Perplexity=15.43, Bits=42026.033.6116.26
BBQParams=95M, Tokens=3B, Full Perplexity=24.75, Bits=42026.033.9325.51
BBQParams=125M, Tokens=5B, Full Perplexity=21.51, Bits=42026.033.9322.15
BBQParams=200M, Tokens=10B, Full Perplexity=17.93, Bits=42026.033.9318.79
BBQParams=300M, Tokens=20B, Full Perplexity=15.43, Bits=42026.033.9316.1