LLaMA3+MoSLoRA

Reported on 2 benchmarks across 2 tasks · 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

Question AnsweringonBoolQ
Accuracy· 2024-06-16
74.6
best: 99.87 (Mistral-Nemo 12B (HPT))
Mixture-of-Subspaces in Low-Rank Adaptation arXiv:2406.11909
Sentence CompletiononHellaSwag
Accuracy· 2024-06-16
95
best: 96.1 (CompassMTL 567M with Tailor)
Mixture-of-Subspaces in Low-Rank Adaptation arXiv:2406.11909