Fine tune llama to near perfectly match GPT -3.5
@waitin4agi_
-
LoRA and Small Base Models as Cost-Effective AI Solutions
By
–
Lora and small base models to the rescue In all seriousness it isn’t THAT expensive to get a bunch of A100s..
-
Can an Indian team build GPT-3.5 equivalent LLM
By
–
Do you think an Indian team sitting in India can build an LLM equivalent to GPT-3.5?
-
Vicuna Training Expansion: From 75K to 4M Samples
By
–
vicuna is trained on 75k samples from http://
sharegpt.com
we're going to train it on 4M samples now! -
Open Source AI Model Outperforms ChatGPT in Reasoning Tests
By
–
Ran my general reasoning test on Vicuna-13B. Open source is going to wreck chatGPT shortly lol.
-
Preference for Midjourney output over alternatives
By
–
idk about you but i think the midjourney one looks better
-
Midjourney’s Impressive Quality Showcased in Discord
By
–
Midjourney is just too good now. Randomly saw this on their discord.
-
GPT Can Already Do This From Last Year
By
–
GPT can already do this now: this clip was from last year
-
LeCun’s AI Predictions Proven Wrong by GPT’s Rapid Progress
By
–
This is the part that people should understand carefully. LeCun is a legend but so confidently says “no chance AI can do X”. The very next year, GPT does it, and quite trivially.
— Varun Mayya (@waitin4agi_) 19 mai 2023
Nobody knows 🤷 https://t.co/LFhZxqTkXuThis is the part that people should understand carefully. LeCun is a legend but so confidently says “no chance AI can do X”. The very next year, GPT does it, and quite trivially. Nobody knows