Finn Howell

HealthLeap, previously @robusthq @onemedical @UCBerkeley

[ML @marlouiise]

A new writing tool ✍️

[Finn Howell @FionnualaHowell]

@marlouiise 😍😍

[Paul Kassianik @kass_paul]

🚀 Exciting to present "Tree of Attacks: Jailbreaking Black-Box LLMs Automatically" 📑🌳 Unlock the power of automated jailbreaks using Tree of Attacks with Pruning (TAP). https://t.co/VHOHGmbQcV

[Finn Howell @FionnualaHowell]

Next Tuesday I'll be talking at the LLMs in Production conference along with 40+ others doing cool stuff in the space. My talk will be at 12:50 pst and I'll be speaking about evaluating LLMs!
Register https://t.co/uRDfD0G9IL to join me.

[Hyrum Anderson @drhyrum]

Honored to share a virtual stage with the inimitable @FionnualaHowell as we talk about how to get started with AI Red Teaming, and how to scale via automated testing.

Drawing on lots of lessons learned from @moo_hax, @ram_ssk, @NMspinach, @rdheeko, @garybits, and so many more.

[Finn Howell @FionnualaHowell]

@djsatoda Birks and socks is A game

[Finn Howell @FionnualaHowell]

@AlanCowen Couldn't agree with this more! I find my prompts are beginning to include more of these types of adjectives (especially succinct)

[Shriya Nevatia @shriyanevatia]

Can someone pls build a better goodreads

Just for the sake of humanity

I know it wont be easy to monetize or make venture scale

Just build pls 😭

[Christopher Potts @ChrisGPotts]

According to the first sentence of 100% of the papers I have reviewed this year, large language models have achieved amazing success in recent years. Perhaps we could settle on the abbreviation "Language Models Are Outstanding", so papers could begin "LMAO, but …" to save space.

[Ian Tracey @ian_dot_so]

Oppenheimer didn’t need Jira or a certified scrum master

[Santiago @svpino]

Something that I've had in my mind for a while now:

The honeymoon phase with Large Language Models is over. People moved away from cool demos to building actual applications.

Our tolerance for mistakes has changed drastically and it will continue getting lower.

[Yann LeCun @ylecun]

This is huge: Llama-v2 is open source, with a license that authorizes commercial use!

This is going to change the landscape of the LLM market.
Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers

Pretrained and fine-tuned models are available with 7B, 13B and 70B parameters.

Llama-2 website: https://t.co/PKrrXgHdem
Llama-2 paper: https://t.co/aINNrXNhMb

A number of personalities from industry and academia have endorsed our open source approach: https://t.co/N7HwgW9Suh

[Finn Howell @FionnualaHowell]

Healthcare often aims for 'normality' over optimization. While reaching 'normal' can benefit much of the population, I would love for a doctor to be willing to go beyond that.

[ali @alitaylor]

i want to go to a doctor that will run every single test possible and give me a personalized plan for optimal health

[Finn Howell @FionnualaHowell]

@Oshi_LLC Yesssss

[Finn Howell @FionnualaHowell]

Dipsea trail never disappoints 😍

[Scott Condron @_ScottCondron]

Building a classifier in 2023

Use @OpenAI's new function calling API to define the possible outputs and then use the "input" argument it returns as the classification

[Finn Howell @FionnualaHowell]

2. The door is locked when it was supposed to be unlocked, host is not answering, airbnb support tells me to go to a coffee shop and use a $25 voucher while they sort it out… it is 11pm and I am alone in a new town. I have to book a hotel, something they can’t help with.

[Finn Howell @FionnualaHowell]

Have had two terrible @Airbnb check in experiences the past few weeks. Especially when traveling alone, opting for the now luxurious front desk hotel experience next time

[Finn Howell @FionnualaHowell]

1. Got into the Airbnb and find someone else’s stuff. After back and forth with the host we realize she gave me the wrong unit, and uses the same door code for every unit 🤯

[Finn Howell @FionnualaHowell]

@eena_kosik @kibm_ucsd Congrats! This is incredible 🙌🏻