David Foster @davidADSP

Author of Generative Deep Learning: Teaching Machines how to Paint, Write, Compose and Play (O'Reilly), #generativeAI, Founding Partner of ADSP. adsp.ai London, UK Joined July 2019

Tweets

419
Followers

764
Following

577
Likes

1K

David Foster @davidADSP

a year ago

@mattshumer_ Yeah looks awesome - any idea how they calculated the $0.19-$0.49 PPM tokens? They say it's based on $2/hour H100 cost and serve rate of 0.03 ms / token I think?

0 0 1 116 0

View Details

David Foster @davidADSP

2 years ago

@Thom_Wolf It's a reference to the fact that an ensemble of all submissions would have scored 81% on the private test set (i.e. 19% of solutions were unsolved by any solution) x.com/fchollet/statu…

Does this mean the ARC-AGI benchmark has saturated? Yes -- the v1 version of the benchmark is starting to saturate. There were already signs of this in the Kaggle competition this year -- an ensemble of all submissions would score 81%. The competition next year will run on

13 33 723 155K 66

0 0 3 506 0

View Details

David Foster @davidADSP

2 years ago

@arcprize @OpenAI o3 solves ARC-AGI? Huuuuge news if that's it...

0 0 4 492 0

View Details

David Foster @davidADSP

2 years ago

@fchollet Out of interest @fchollet, what % of arc test set puzzles remain unsolved by any submitted solution? And what would the top 2 entries score if ensembled (I know this means they'd have 4 attempts). Just curious how much they overlap.

0 0 1 105 0

View Details

David Foster @davidADSP

2 years ago

@OfficialLoganK What's the rate limit?

0 0 0 89 0

View Details

David Foster @davidADSP

2 years ago

@jsuarez @hirschibar Awesome write up! What about action masking - i.e. how do you handle cases where certain actions aren't possible (and the env returns you the mask at each timestep). Is this something PufferLib supports?

2 0 1 57 0

View Details

David Foster @davidADSP

2 years ago

@lmarena_ai @01AI_Yi Will the multimodal Llama 3.2 models be added to the overall leaderboard?

0 0 0 636 0

View Details

David Foster @davidADSP

2 years ago

@lmarena_ai @lmsysorg Will Llama 3.2 (the multimodal models) and Gemini 1.5-002 be added to the main leaderboard?

0 0 0 1K 0

View Details

David Foster @davidADSP

2 years ago

@SullyOmarr Would you be willing to share the leaderboard from your evals?

1 0 1 267 0

View Details

David Foster @davidADSP

2 years ago

Spot the data viz fail 🤦‍♂️@BBC @BBCPolitics @BBCNews

0 0 1 321 0

View Details

David Foster @davidADSP

2 years ago

@giffmana @pastaraspberry What is it about Gemma that makes it open, but not open source? Thanks!

1 0 0 312 0

View Details

David Foster @davidADSP

2 years ago

@giffmana @pastaraspberry How are you defining open vs open source. Thanks!

1 0 1 341 0

View Details

David Foster @davidADSP

2 years ago

@NPCollapse Funny story - William Peebles co-authored the Mar 2023 Diffusion Transformer paper on which Sora is based, whilst at Meta as an intern. But then joined OpenAI last year to co-lead Sora. So I guess they did know how to do it, but let him leave 😂

0 0 3 644 0

View Details

David Foster @davidADSP

3 years ago

@Thom_Wolf Theory of Everything: youtube.com/watch?v=q7i_DY…

0 0 1 65 0

View Details

David Foster @davidADSP

3 years ago

@realGeorgeHotz Given the current breakthroughs, "linguistics" is a left-field candidate 🤔

0 0 0 162 0

View Details

David Foster @davidADSP

3 years ago

@nickfloats Does the --iw parameter affect remixes? In the docs it says it doesn't, but I'm never sure how much to trist the docs :)

0 0 1 103 0

View Details

David Foster @davidADSP

3 years ago

@nickfloats Related question / challenge - how do you get Midjourney to output the usual meaning of 'fork in the road', rather than this? Changing the prompt to use different words isn't allowed 😃