GALI(Llama3-8b-ins-8k-to-32k)

Reported on 2 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing2 results

Long-Context UnderstandingonLongBench
Average Score· 2025-02-04
45.38
best: 46.22 (GALI(Llama3-8b-ins-4k-to-16k))
A Training-Free Length Extrapolation Approach for LLMs: Greedy Attention Logit Interpolation (GALI)arXiv:2502.02659
Long-Context UnderstandingonL-Eval
Average Score· 2025-02-04
42.79
best: 59.21 (GALI(Llama3-8b-ins-4k-to-16k))
A Training-Free Length Extrapolation Approach for LLMs: Greedy Attention Logit Interpolation (GALI)arXiv:2502.02659