Harish Vadlamani

Qui vivra verra!

[Fabrizio Romano @FabrizioRomano]

Cody Gakpo. Had to be him. 🥺

[Philipp Schmid @_philschmid]

Yesterday Apple strongly advocated for on-device AI with @Apple Intelligence! Today, we know they will run ~3B LLM on device (Mac, iPhone, iPad) using fine-tuned LoRA Adapters for different tasks, claiming to outperform other 7B and 3B LLMs! 👀

What we know:

📱 On-device 3B with grouped-query-attention, activation and embedding quantization running on the neural engine

📊 iPhone 15 Pro 0.6 ms time-to-first-token with 30 tokens/second latency

❌ Server model size and details are unknown

🔄 Dynamically load, cache, and swap LoRA adapter models as needed

📈 On-device model 49K vocab size, while the server model 100K

🛠️ Used Rejection Sampling and descent policy optimization and a leave-one-out advantage for RLHF

📝 Used synthetic data generation for tasks like summaries

🔍 750 evaluation samples for each production use case

🇺🇸 Seems to be an English model, this would explain the limitation to US only

🧠 Used Apple's AXLearn framework (JAX) and FSP to train on TPUs and GPUs

⚡ 3B + Adapter outperforms Phi-3 mini on summarization

🏆 3B + Adapter achieves 78.7% on IFEval > Mistral 7B; Server Model matches GPT-4-Turbo

[Liverpool FC @LFC]

WE ARE THE CARABAO CUP WINNERS!!!!!!! 😍

[Jamie Carragher @Carra23]

When it takes this long for an answer from @FA_PGMOL over a contentious decision you know there’s been a f**k up!

[Migs @not_migwell]

@ChantLFC

[LFC Transfer Room @LFCTransferRoom]

.@NUFC 🤫

[Samuel @SamueILFC]

Fabrizio Romano updates on Moises Caicedo every hour.

[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

[Harish Vadlamani @harish3110]

@_pauljoyce

[Harish Vadlamani @harish3110]

@Angelina_Magr @fastdotai Hi Angelina, would you also consider candidates from Canada? @Angelina_Magr

[François Chollet @fchollet]

PyTorch-Lightning is what you get when you order Keras from AliExpress

[Jay Alammar @JayAlammar]

The Illustrated Stable Diffusion

https://t.co/sbjKHP4Pol

New post!

Over 30 visuals explaining how Stable Diffusion works (diffusion, latent diffusion, CLIP, and a lot more).

[Jarno Opmeer @jarno_opmeer]

Giving away 3 x F1 22 Champions edition codes here on twitter.
- Follow, like, RT to enter

[Marques Brownlee @MKBHD]

I’m gonna pack up this @ColorWare Nintendo Switch OLED and ship it to 1 random follower who RTs this tweet in the next 24 hours. Good luck!

[LeBron James @KingJames]

PARIS HERE WE COME!!!!!!!! @LFC 👏🏾👏🏾👏🏾👏🏾👏🏾👏🏾🔴🔴🔴🔴🔴🔴

[Jeremy Howard @jeremyphoward]

There's a interesting new NLP competition at Kaggle about patent concept similarity. The dataset is easy to work with, and there's lots of room for creative approaches.

I've provided a starter notebook, which is also an intro to @huggingface Transformers
https://t.co/SC4dR3BcYm

[Harish Vadlamani @harish3110]

@TheZachMueller Cmd + / (Window split)
Ctrl + ~ (Open terminal)
Fn + F1 (Command Palette?)

[Harish Vadlamani @harish3110]

And I just got robbed an hour of precious sleep! :/

[Ryan @ryan3levis]

So if Liverpool make it to the CL quarters and the FA Cup semis, drawing City in both... our April could be:

5/6 April - Man City - CL
10 April - Man City - PL
12/13 April - Man City - CL
16/17 April - Man City - FA Cup
19/20 April - Man United - PL
23/24 April - Everton - PL