Louis Hernandez

2 posts

Louis Hernandez

Louis Hernandez

@Louis_Hdez

Katılım Ekim 2014
841 Takip Edilen22 Takipçiler
Louis Hernandez retweetledi
Koubbi Hugo
Koubbi Hugo@HugoKoubbi·
The LoRA algorithm seems to be the most widespread fine-tuning method for LLMs. LoRA reduces parameter count by employing low-rank matrix factorization on attention mechanisms. In arxiv.org/abs/2402.15415 w/ @m_boussard and @Louis_Hdez, we study the impact of LoRA on tokens 1/10
Koubbi Hugo tweet media
English
1
1
5
693