Pith. sign in

Title resolution pending

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

eess.SP 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Power-of-Two (PoT) Weights in Large Language Models (LLMs)

eess.SP · 2025-05-31 · conditional · novelty 4.0

Power-of-two weight quantization applied post-training to a 124M GPT-2 model degrades cross-entropy from 3.17 to about 4.1-4.5 at 4-6 bits, while promising memory and bit-shift savings.

citing papers explorer

Showing 1 of 1 citing paper.

  • Power-of-Two (PoT) Weights in Large Language Models (LLMs) eess.SP · 2025-05-31 · conditional · none · ref 3

    Power-of-two weight quantization applied post-training to a 124M GPT-2 model degrades cross-entropy from 3.17 to about 4.1-4.5 at 4-6 bits, while promising memory and bit-shift savings.