Joel Kronander (@jkronand) / X

Joel Kronander

1,527 posts

Joel Kronander

@jkronand

I try to learn something every day✨Former Head of ML at Nines, Former Head of Synth data at Scale AI 💫

Palo Alto, CA

Joined March 2008

Joel Kronander
@jkronand
Apr 4, 2023
I recently demonstrated GPT4 to my spouse's 101-year-old grandfather, who remains in excellent health and has a sharp mind. Following my demonstration, he paused thoughtfully and then said something I will remember — “This technology instills hope for our future. It's high time
498K
Joel Kronander
@jkronand
Sep 24, 2023
Stanfords DSPy is the best high level LLM programing framework I have seen this far. Langchain never resonated with me; despite being an early LLM framework, its design and abstractions felt overly complex. DSPy, on the other hand, is a huge step in the right direction. DSPy
129K
Joel Kronander
@jkronand
Mar 30, 2023
An interesting new Nature paper compares fMRI recordings with activations across layers in a language model, and find evidence of correlations. The study seems to suggests that brain regions located at the top of the language hierarchy, responsible for nature.com/articles/s4156…
145K
Joel Kronander
@jkronand
May 6, 2023
Six years ago, Geoffrey Hinton asserted that AI would take over radiology within five years, suggesting we cease training radiologists. Was he correct? The situation is more complex than simply being right or wrong. While AI has surpassed radiologists in certain diagnostic
387K
Joel Kronander
@jkronand
Feb 9, 2023
Deep learning is typically bottlenecked by memory not compute ⚡️Flash Attention ⚡️ optimizes transformers, like GPT, to minimize costly GPU memory fetches and achieves impressive speedups of 2-4x, 5-20x less memory intensive, and enables scaling to longer arxiv.org/abs/2205.14135
31K
Joel Kronander
@jkronand
Jan 29, 2023
Self-consistency is underrated for improving accuracy for LLMs in a range of reasoning and arithmetic tasks. It works with any off-the-shelf LLM, eg GPT3 variants, and also provides estimates of how certain the LLM is of the provided answer. arxiv.org/abs/2203.11171 Takeaways👇
57K
Joel Kronander
@jkronand
Apr 21, 2023
A simple trick to make LLMs “calibrated” — ie “to know when it doesn’t know something” — is to reformulate the answers as a single word or a short phrase, and look at the predicted logprobs of the word. As LLMs are trained to predict the probability of the next token, they are
46K
Joel Kronander
@jkronand
Feb 4, 2023
🤖️LLM can self-improve 🧠 1) Self-consistency boosts reasoning skills by sampling multiple paths & finding the most consistent answer But more samples = more comp. requirements. 💻 2) but we can train better LLM with self-generated solutions from 1) arxiv.org/abs/2210.11610
33K
Joel Kronander
@jkronand
Mar 30, 2023
What it you had trained a model to play legal moves in Othello by predicting the next move, and found that it had spontaneously learned to compute/represent the full board state in it's weights - an emergent world representation? That's just what this thegradient.pub/othello/
179K
Joel Kronander
@jkronand
Apr 3, 2023
Insightful paper that succinctly covers essential high-level knowledge to keep in mind regarding LLMs: - Large language models (LLMs) predictably improve with increasing investment, but many key behaviors emerge unpredictably. - LLMs often learn and use representations of the
38K
Joel Kronander
@jkronand
Feb 1, 2023
✨Neat LLM trick for 📈 math & logical abilities ✨ Improves on Chain of Thought (CoT) prompting by 1) Replace natural language, step by step instructions, in CoT examples with commented, stepwise, python code. 2) Run the code Several recent papers on this (see refs below⬇️)
11K
Joel Kronander
@jkronand
Jan 24, 2023
Want to know a simple trick for LLMs to generate more plausible long documents, breaks out of repetition better, and more reasonably truncate low probability tokens? Learn about LLM truncation sampling! Some takeaways from arxiv.org/abs/2210.15191 👇🧵
17K
Joel Kronander
@jkronand
Sep 26, 2023
Replying to @ylecun
GPT4 would never make that mistake.
17K
Joel Kronander
@jkronand
Feb 4, 2023
LLMs suffer from overconfidence and poorly calibrated uncertainty estimates However, self-consistency, where on samples multiple paths & finds the most consistent answer, seems to offer a practical solution. Interesting figure from page 4 in "LLMs can self improve" paper
19K