damn my bad man

We asked our ChainGPT Chat Bot how LLMs work:

'An LLM is a robot who reads a lot and learns how humans talk. It figures out how words fit together and can create its own sentences that sound real.'

How would you add to this? 🤔

I used the chatbot arena data from lmsys.org to create a visualization of LLM’s Elo rating changes. You can see:

1. The gap between various companies/open source projects is narrowing.
2. The major players are gradually becoming the various big tech companies.

📌 Pretty interesting proposal in this paper - Finetuning an LLM without actually training its own weights.

📌 'Tuning Language Models by Proxy'

🔥 Proxy-Tuning relies on a setup where you have a large LLM that you don't want to/can't fine-tune, and a pair of small LLMs that…

güzel bir ide'ye denk geldim demin: zerve ai

llm ve datasetleri sürükle bırak yaparak kolayca fine-tune edebiliyorsunuz.

huggingface ve bedrock entegrasyonu da varmış. yani modelleri oradan çekebilir veya host edebiliyormuşsunuz.

Kyle rayner ^_^

Simplest explanation of LLM by Andrej Karpathy

LLM fine-tune etmeyi aşırı kolay hale getiren muhteşem bir araç:

github.com/hiyouga/LLaMA-…

The jaykyle fandom is dead! Repost to keep it DEAD.

エーベンホルツさん落書き

ben işe başladığımdan beri sektör çok değişti. genAI ve LLM'ler hakkında fikir sahibi olmak gerekiyor. bir reddit kullanıcısı 40'tan fazla medium seniority Data Science pozisyonuna bakarak bu ilanların gerekliliklerinden bir yol haritası çıkartmış. iş arayanlar için faydalı…

『EasyNovelAssistant』
軽量で規制も検閲もない日本語ローカル LLM『LightChatAssistant-TypeB』で簡単に文章生成

ローカル特権の永続生成 Generate forever で、当たりガチャを積み上げます。
1ステップ導入で簡単にサンプル群をお試し。

要 RAM 16GB WinPC, Geforce 4GB
github.com/Zuntan03/EasyN…

LLB✅,AHCZ✅,25/04/24 LLM✅… #masteredit

I studied human rights law (what a waste of my time) but seriously considering going back to law school for a LLM in Media Law, Privacy and defamation so I can take all these ghouls to court & leave them with only a Kleenex to wipe their tears.

Unpopular opinion:
LLM’s are coming for podcast.

The next few iterations of gpt will become better at summarizing and conveying information — better than any podcasts.

jungwon cowboy hat pt 2

#FATE_PLUS_IN_OAKLAND #FATE_PLUS_IN_US

Llama 3 just changed the LLM game.

People are finding wild use cases at GPT-4 level. There is a massive movement in the open source community.

10 examples (and ways to use Llama 3):

Old friend

Lastnight Was One Of Them Nights🩷 #LLM