sebhtml Profile Banner
Sébastien Boisvert Profile
Sébastien Boisvert

@sebhtml

Followers
1K
Following
229
Media
167
Statuses
6K

AI, C++, Rust, Pytorch, Software Developer, Calisthenics, Squat, Deadlift

Québec, Canada
Joined December 2009
Don't wanna be here? Send us removal request.
@sebhtml
Sébastien Boisvert
1 year
ARC Prize 2024 Puzzle 0ca9ddb6 https://t.co/RB6UEwmvj8 via @YouTube
0
0
1
@sebhtml
Sébastien Boisvert
1 year
Rich Sutton’s new path for AI | Approximately Correct Podcast https://t.co/kGuO9HslQL via @YouTube
0
0
1
@SquatUniversity
Squat University
1 year
Big Shoe Companies LIED TO US!🤯
97
1K
12K
@ylecun
Yann LeCun
8 years
Training with large minibatches is bad for your health. More importantly, it's bad for your test error. Friends dont let friends use minibatches larger than 32.
Tweet card summary image
arxiv.org
Modern deep neural network training is typically based on mini-batch stochastic gradient optimization. While the use of large mini-batches increases the available computational parallelism, small...
26
498
1K
@yacineMTB
kache
1 year
Canada should revoke access to our free healthcare to non citizens If citizens can't get family doctors, why are we giving access to so called "asylum" seekers?
113
109
2K
@sebhtml
Sébastien Boisvert
1 year
I moved the Dropout unit at the correct place.
0
0
0
@sebhtml
Sébastien Boisvert
1 year
0
0
1
@sebhtml
Sébastien Boisvert
1 year
Learning to solve the example 1 of puzzle 3aa6fb7a in the ARC prize @arcprize
0
0
1
@sebhtml
Sébastien Boisvert
1 year
For the ARC price https://t.co/ma3aNrPo5P, today I have increased the number of layers from 4 to 8. To do this on the NVIDIA A40 GPU on https://t.co/sIxaPc8nFn, I changed my batch_size from 1024 to 512 and changed num_layers from 4 to 8. #AI @arcprize
Tweet card summary image
arcprize.org
ARC Prize is a $1,000,000+ nonprofit, public competition to beat and open source a solution to the ARC-AGI benchmark.
0
0
1
@sebhtml
Sébastien Boisvert
1 year
For the ARC price https://t.co/ma3aNrPo5P, today I fixed a bug that caused my neural network to not see all the input text of a sample because the context was 140 characters but the samples were 211 characters
0
0
0
@karpathy
Andrej Karpathy
1 year
You can tell the RL is done properly when the models cease to speak English in their chain of thought
293
387
7K
@sebhtml
Sébastien Boisvert
1 year
I made a meme image. #AI #LLMs #annoyedbird
0
0
0
@EmmanuelMacron
Emmanuel Macron
5 years
Vive la République, vive la France !
4K
4K
44K
@sebhtml
Sébastien Boisvert
1 year
I am using @runpod_io to train my AI for the @arcprize I think that I should have eventually something working for puzzle 3aa6fb7a using pytorch and xformers and transformers. hehe #AI #puzzles
0
0
2
@vsbuffalo
Vince Buffalo
1 year
Rust is such a beautiful language for working with genomic data. Clean syntax, functional programming, and extremely fast.
8
10
165
@runews
Russian Market
1 year
Disaster in France
330
5K
31K
@minchoi
Min Choi
1 year
892
4K
38K
@elonmusk
Elon Musk
1 year
🇺🇸🇺🇸 Happy Independence Day! 🇺🇸🇺🇸
16K
29K
413K