old pepper 🌶️

58 posts

old pepper 🌶️

old pepper 🌶️

@oldpepper12

prev. founding engineer @edendotso | opinions that of claude code

Toronto Katılım Şubat 2019
38 Takip Edilen11 Takipçiler
old pepper 🌶️ retweetledi
jacob paris ▲
jacob paris ▲@jacobmparis·
VICTORY! GitHub officially back to 5 nines of uptime
jacob paris ▲ tweet media
English
13
56
2.3K
63.7K
old pepper 🌶️ retweetledi
old pepper 🌶️ retweetledi
Andrej Karpathy
Andrej Karpathy@karpathy·
- Drafted a blog post - Used an LLM to meticulously improve the argument over 4 hours. - Wow, feeling great, it’s so convincing! - Fun idea let’s ask it to argue the opposite. - LLM demolishes the entire argument and convinces me that the opposite is in fact true. - lol The LLMs may elicit an opinion when asked but are extremely competent in arguing almost any direction. This is actually super useful as a tool for forming your own opinions, just make sure to ask different directions and be careful with the sycophancy.
English
1.7K
2.4K
31.1K
3.3M
old pepper 🌶️ retweetledi
Fleetwood
Fleetwood@fleetwood___·
Fleetwood tweet media
ZXX
23
126
946
58.5K
old pepper 🌶️ retweetledi
Weights & Biases
Weights & Biases@wandb·
BREAKING: Meek Mill has abandoned the W&B web dashboard entirely. Says he only monitors training runs in the terminal now. "I don't need a browser to know loss is converging," sources close to the rapper confirm.
Weights & Biases tweet mediaWeights & Biases tweet media
English
36
103
1.9K
106.9K
old pepper 🌶️ retweetledi
Terrible Maps
Terrible Maps@TerribleMaps·
Believe it or not, Germany’s 5 largest cities lie perfectly on a 4th-degree polynomial
Terrible Maps tweet media
English
383
1.1K
30.4K
5M
old pepper 🌶️ retweetledi
hayden
hayden@haydendevs·
"UWaterloo preferred" in the job posting
hayden tweet media
English
7
2
173
7.3K
old pepper 🌶️
old pepper 🌶️@oldpepper12·
"we found a task where LLMs struggle massively! write a C compiler in Rust and they'd work great. have them formulate the same problem as a tuple M=(Q,Γ,Σ,δ0,F) satisfying δ:(Q\F) X Γ =? Q x Γ x {L, R} and their zero-shot performance drops to 0!"
Paras Chopra@paraschopra

We found a task where LLMs struggle massively! Give them a coding problem in Python and they'd work great. Give the same problem in brainfuck and zero-shot their performance is ~0% +[--------->+<]>+.++[--->++<]>+.

English
0
0
1
83
old pepper 🌶️
old pepper 🌶️@oldpepper12·
look at my oss dawg
tldraw@tldraw

Welcome @BlackRock as the latest customer of the tldraw SDK. This deal brings the total assets under management (AUM) of tldraw's customers to just over $14 trillion.

English
0
0
0
30