Open-LLaMA-3B-v2

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 AnsweringonPIQA
Accuracy· 2023-10-10
76.2
best: 90.1 (Unicorn 11B (fine-tuned))
Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning arXiv:2310.06694
Sentence CompletiononHellaSwag
Accuracy· 2023-10-10
67.6
best: 96.1 (CompassMTL 567M with Tailor)
Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning arXiv:2310.06694