Testing good-gpt-2 and also-good-gpt2 in chatbot arena

This video is a series of experiments using good-gpt-2 and also-good-gpt2 chatbots in lmsys arena in battle mode
1000x MasterClass: / 1000x-dev-103326330
Streamlit course: / exclusive-course-10275...
Fastapi course: / learn-fastapi-26-95041684
Auto Streamer: www.autostreamer.live/
Search 200+ echohive videos and code download links:www.echohive.live/
Chat with us on Discord: / discord
Follow on twitter(X) : / hive_echo
CHAPTERS:
00:00 lightbulb css/js animation
02:25 spinning circles animation
02:56 My patreon benefits
03:30 spinning circles continued
06:19 Beautiful webpages
#goodgpt2 #gpt2 #lmsys #chatbot #chatbotarena #largelanguagemodels #artificialintelligence #pythonprogramming #pythonprojects #pythonai

[@ryzikx]

pretty big step up from gpt-4 if its really the size of gpt-2

[@echohive]

Yeah I would agree. But why would they name it gpt2 otherwise.

[@therainman7777]

It’s not GPT-2, nor is it some other model that’s the same size as GPT-2. Trust me, that is not possible.

[@echohive]

1000x MasterClass: www.patreon.com/posts/1000x-dev-103326330
Streamlit course: www.patreon.com/posts/exclusive-course-102751857
Fastapi course: www.patreon.com/posts/learn-fastapi-26-95041684
Auto Streamer: www.autostreamer.live/
Search 200+ echohive videos and code download links:www.echohive.live/
Chat with us on Discord: discord.gg/PPxTP3Cs3G
Follow on twitter(X) : twitter.com/hive_echo

[@nabil2272]

how is thw best good-gpt-2 or also-good-gpt2 ? plz im not good

[@echohive]

I’m am not sure which one is better

[@dgfilms1712]

gpt2 isnt avaible, how did you get it to show up in the areana dropdown

[@drlordbasil]

its only again eachother.

[@drlordbasil]

and is random

[@echohive]

You have to find them in the arena(battle) mode

[@ScottzPlaylists]

I hope OpenAI goes completely Open Source including Training Methods and Code, Training DB, Inference Code that runs on 1 to 1M GPUs. The OSS community will advance it faster than OpenAI could. ❗❗❗❗❗
GPT-3 would be a good start.👍
They could plop their latest training run into the OSS code and still keep their precious Moat (advantage) ❗❗❗❗
When 4.5 or 5 comes out, they could release GPT 3.5 weights. I think it would be better for them in the long run. You❓❓❓

[@AI.24.7]

They will have to go OSS if they want to keep ahead of Meta

[@echohive]

These maybe small open source models they are testing cuz of the name gpt2

[@AGIBreakout]

🤯 Great Idea... I agree. we'll see.❗

[@user-en4ek6xt6w]

Impossible read their last article they say no open-source lol

[@echohive]

Oh I didn’t know that :(