據悉,亞馬遜現有比GPT-3.5高出16%性能的語言模型,且是公開了其源代碼供行業內人士及大眾參考學習的,唯一不足的地方是其模型訓練參數只有10億個。

其研究、代碼和數據:
1、學習:https://lnkd.in/g2Xn4462
2、代碼:https://lnkd.in/gJ7BxVJJ
3、資料集:https://lnkd.in/gdp6NF9k

領英貼文鏈接:https://www.linkedin.com/posts..../yablonassaf_gpt-cot

#AmazonScience #LLM #AI #NLP #GPT

Assaf Yablon on LinkedIn: #gpt #cot #amazonscience #llm #ai #nlp | 126 comments

Amazon outperforms GPT-3.5 by 16%! . . Amazons model, trained with 1 billion parameters, has surpassed the previous state-of-the-art LLM (#GPT-3.5) by 16%… | 126 comments on LinkedIn
imageimage