LLMS

→ View original post on X — @kirkdborne

New Fine-Tuning Method Improves LLM Performance

By

@kirkdborne

–

22 February 2024 4h10

RT from @bindureddy at @abacusai >> New Way To Fine Tune #LLMs For Better Performance! The research paper: https://
arxiv.org/abs/2402.13228 The model: https://
huggingface.co/abacusai/Smaug
-Mixtral-v0.1
…

→ View original post on X — @jxmnop

Google researcher struggles to beat Mistral 7B model performance

By

@jxmnop

–

22 February 2024 3h12

> be google researcher
> exec says we need to beat mistral 7b
> “reclaim the narrative"
> train 7b model
> run eval. check plots
> cant do it. maybe with 8b
> exec says no dice
> “we'll look like chumps"
> ok
> maybe we can tie embeddings and linear_f, save some params
> not

Fine-tuned Small Open-Source Models Represent the Future

By

–

22 February 2024 2h42

Thanks for highlighting our work. The future is fine-tuned small open-source models!

2B and 7B Models Now Available in Free Trial

By

–

22 February 2024 1h57

Both 2B and 7B are available in our free trial!

→ View original post on X — @skirano

Gemini 1.5 Pro Analyzes Tokenization Video Content

By

@skirano

–

22 February 2024 1h56

Gemini 1.5 Pro is super impressive. I am feeding it @karpathy
's new lesson on Tokenization, and I can easily: – Summarize the video.
– Identify the most important concepts.
– List action items for further exploration. I broke the video down into 30-minute chunks.

Gemma-7B Now Available as Serverless LLM Endpoint on Predibase

By

–

22 February 2024 1h50

🚀 #Gemma-7B now available for prompting in Predibase as a serverless #LLM endpoint.

Try it out for free: https://t.co/HkWmdZvJFT pic.twitter.com/KRt5oCCnHP
— Predibase by Rubrik (@predibase) 22 février 2024

#Gemma-7B now available for prompting in Predibase as a serverless #LLM endpoint. Try it out for free: https://
pbase.ai/3IakXUk

→ View original post on X — @perplexity_ai

Perplexity Labs Introduces Gemma 2B and 7B Models

By

@perplexity_ai

–

22 February 2024 0h49

Introducing new additions to Perplexity Labs: Experience Gemma 2B and 7B models known for impressive performance despite being lightweight. Try it now on https://t.co/2yYQB9jln7. pic.twitter.com/PIipmESdXN
— Perplexity (@perplexity_ai) 21 février 2024

Introducing new additions to Perplexity Labs: Experience Gemma 2B and 7B models known for impressive performance despite being lightweight. Try it now on http://
labs.pplx.ai.

Predibase LoRA Land democratizes AI for smaller companies

By

–

21 February 2024 22h00

Nice recap from #Datanami: "The #efficiency and affordability offered by Predibase [LoRA Land] levels the playing field for smaller companies in the AI race. Its a game changer for companies that want to deploy a wide range of #specialized LLMs."