倢愷 Oscar
Jun 5, 2024

感謝感謝XD

我自己是習慣先用小規模的lora來確定lora的上限,畢竟多個epoch跟增多data能帶來的效果都能夠提前預測出來,或是最少能建立一個基礎的認識。

不過更常還是有點依賴研究員的直覺,包含我們這個專案手中能收到的diverse instruction tuning data有多大,以及這個能力有多偏離原本的base model,都會影響我的決策XD,但是確實很多時候都沒有辦法100%驗證決策。

其實這也是我認為大模型開發上最難的一點,就是學習成本太高,難以快速累積出經驗法則,底層研究員都難以累積經驗了,更遑論上層人要累積出決策的經驗。

Sign up to discover human stories that deepen your understanding of the world.

Free

Distraction-free reading. No ads.

Organize your knowledge with lists and highlights.

Tell your story. Find your audience.

Membership

Read member-only stories

Support writers you read most

Earn money for your writing

Listen to audio narrations

Read offline with the Medium app

倢愷 Oscar
倢愷 Oscar

Written by 倢愷 Oscar

我是倢愷,CTO at TeraThinker an AI Adaptive Learning System Company。AI/HCI研究者,超過100場的ML、DL演講、workshop經驗。主要學習如何將AI落地於業界。 有家教、演講合作,可以email跟我聯絡:axk51013@gmail.com

Responses (1)

Write a response