Latent.Space @latentspacepod Twitter profile

Last Seen Profiles

@TuNa__fn

@urkumaresanpro

@salib_iraq

@mmdjml671854822

@nbcgameofgames

@NFT_ASCII

@bokeplokalmalam

@thor_com_se

@CryptoSyzygy1

@onlyseegaylight

@KatleenWils

@wonsternbin

@jandakembangstw

@Alwlh07Qlm

@jezzylw

@IM_Communityitw

@Foreverwith9you

@CoachRich_G

@YOUNGKD213

@Habigin

@jandakembangstw

@MinOCW

@E_Matsu_82

@Ahmed_Rm_

@megxharry

@HertsSheriff

@J5daigada

@yasminaghajan

@valentimesph

@lindow

@aysha_2022

@goneroty

@stwmaniax

@kokiandrori

@bokeplokalmalam

@Alpha_Femalee

Latent.Space

@latentspacepod

22 days

Some very promising arena results, congrats @xai & @elonmusk ! Open invitation for a grok-2-large release podcast with us, people would love to hear the technical details 👀

xAI

@xai

22 days

1K

2K

9K

4

13

311

Latent.Space

@latentspacepod

9 months

🆕 NeurIPS 2023 Recap — Best Papers! Our selection for AI Engineers this year includes: - Word2Vec ( @JeffDean ) - DPO ( @rm_rafailov ) - Datablations ( @Muennighoff ) - QLora ( @Tim_Dettmers ) - DataComp ( @sy_gadre ) - LLaVA ( @imhaotian ) - Tree of Thoughts

NeurIPS 2023 Recap — Best Papers

Our selection for AI Engineers: Word2Vec (with Jeff Dean), State Space Models (with Chris Ré), Emergence Mirage, DPO, Datablations, QLora, LlaVA, DataComp, Tree of Thought, CogEval, Voyager

www.latent.space

4

71

301

Latent.Space

@latentspacepod

11 months

What's the #1 trick to massively stand out in your career? @jeremyphoward drops his #1 tip. The bar has been set

6

40

192

Latent.Space

@latentspacepod

10 months

🆕 The New Kings of Open Source AI Recapping how @MistralAI took over Open Source AI, how the definition of Open is evolving, and why engineers freaked out about Copilot losses at @aidotengineer Memes of the month: @itsmingjie @nearcyan ft @TheNoahHein

3

22

171

Latent.Space

@latentspacepod

5 months

🆕 the one and only @jxnlco on: • Why Pydantic is -still- All You Need • Why most AI frameworks aren't VC backable • Why you'd be happier being bootstrapped • Why you should NOT hire ML Engineers • the definitive origin of Instructor (and bonus

High Agency Pydantic > VC Backed Frameworks — with Jason Liu of Instructor

Structured output vs function calling, state of the AI engineering stack, when to NOT raise venture capital, and career advice for aspiring AI Engineers

www.latent.space

3

16

130

Latent.Space

@latentspacepod

2 months

🆕 pod: The Yolo Researcher Metagame with @YiTayML ! OpenAI (ca. GPT4): ~600 people Google Gemini: ~950 coauthors @RekaAILabs : 20 people @sama once speculated on the qualities of "10,000x AI researchers", and more recently @_jasonwei described the "Yolo

5

19

120

Latent.Space

@latentspacepod

9 months

🆕 The Busy Person's Intro to Finetuning & Open Source AI with @winglian of Axolotl Covering the SF AI meetup with @NousResearch , @Teknium1 , and all the required knowledge to get started navigating Open Source AI and finetuning them. (special cohost

The Busy Person's Intro to Finetuning & Open Source AI - Wing Lian, Axolotl

On Teknium and the Nous Research SF meetup, creating the most popular finetuning library, and taking research from papers to tools - from QLoRA to Mamba State Space Models - in days.

www.latent.space

2

24

106

Latent.Space

@latentspacepod

8 months

Prediction: "In the next 2 years, every LLM will become LMMs" ( @chipro 's term for Large Multimodal Models) 🆕: How to train your own Large Multimodal Model with @HugoLaurencon and @LeoTronchon of @Huggingface ! We follow how HuggingFace's M4 team

1

19

73

Latent.Space

@latentspacepod

1 month

🆕 pod with @ThomasScialom of @AIatMeta ! Llama 2, 3 & 4: Synthetic Data, RLHF, Agents on the path to Open Source AGI shoutouts: - Why @ylecun 's Galactica Instruct would have solved @giffmana 's Citations Generator - Beyond Chinchilla-Optimal: 100x

Llama 2, 3 & 4: Synthetic Data, RLHF, Agents on the path to Open Source AGI

Llama 2 lead and Llama 3 post-training lead Thomas Scialom of Meta/FAIR, on the Chinchilla trap, why Synthetic Data and RLHF works, and how Llama4's focus on Agents will lead us to Open Source AGI.

www.latent.space

4

16

64

Latent.Space

@latentspacepod

8 months

🆕 The Accidental AI Canvas The full story of how @steveruizok took @gdb 's GPT4V Demo to its logical conclusion, and accidentally invented the future of multimodal prompting with @tldraw . Now on YouTube + wherever fine podcasts are sold.

0

23

57

Latent.Space

@latentspacepod

1 month

🆕 The Winds of AI Winter The vibes have shifted. On @leopoldasch vs Sequoia, Goldman Sachs, @benedictevans , @cpaik 's End of Software and why the AI Engineer rises above it all. The future is here, but it is not evenly distributed.

2

7

55

Latent.Space

@latentspacepod

19 days

🆕 Building AI for The People Never has so much been shipped for so many by so few. @jeremyphoward is back on the pod! sharing the founding journey of @AnswerAI , predicting the @OpenAI governance crisis, hiring 1000x researchers and developers like

AI Magic: Shipping 1000s of successful products with no managers and a team of 12 — Jeremy Howard...

Why Answer is a PBC, going from QLoRA to QDoRA, updating BERT for 2025, creating FastHTML, predicting the OpenAI governance crisis, and a preview of the future of "Dialogue Engineering".

www.latent.space

1

12

54

Latent.Space

@latentspacepod

1 year

🚨 EMERGENCY LATENT SPACE SATURDAY 12PM PT @simonw , @altryne , @FanaHOVA , @swyx Join us to talk how @OpenAI chat interpreter changes coding -and- chat, forever.

1

16

53

Latent.Space

@latentspacepod

3 months

🆕 pod: How to train a Million Context LLM! @ylecun says we should publish, or perish. We asked @markatgradient to spill ALL the beans on how his team extended Llama-3 to 1M+ context with ~perfect @GregKamradt NIAH evals! Also covering: - A Brief

How to train a Million Context LLM — with Mark Huang of Gradient.ai

Scaling Llama3 beyond 1M context window with ~perfect utilization, the difference between ALiBi and RoPE, how to use GPT-4 to create synthetic data for your context extension finetunes, and more!

www.latent.space

2

11

49

Latent.Space

@latentspacepod

8 months

Are GPU clouds intentionally losing money on last month's Mixtral inference price war? Probably. Depends on batching and aggressiveness of accounting. As @NaveenGRao puts it - a lot of people are about to learn the meaning of "TCO"...

swyx.io

@swyx

8 months

🆕 The Four Wars of the AI Stack Our Dec 2023 recap also includes a framework for looking at the key business battlegrounds of all of 2023: In Data: with OpenAI announcing a partnership with Axel Springer (see also its deal with the AP and its Data

13

64

302

6

4

45

Latent.Space

@latentspacepod

9 months

🆕 pod: with @BEBischof of @_hex_tech ! On putting AI Magic into Notebooks, and how RAG is actually a recommendation systems problem. Also an @AAAzzam hot take: LLMOps is more like an "iron mine" than a "gold rush"!

4

11

44

Latent.Space

@latentspacepod

4 months

"You shouldn't even be hiring MLEs" @jxnlco on how companies are unnecessarily hiring MLEs when they should be looking to the under-utilized but motivated engineers that have a desire to pursue AI Engineering.

2

9

45

Latent.Space

@latentspacepod

3 months

🆕 ICLR 2024: Best Papers (Part 1) We present our selections of outstanding papers and talks thematically introducing topics for AI Engineers to track: Section A: ImageGen, Compression, Adversarial Attacks - Auto-encoding Variational Bayes ( @dpkingma ) - Würstchen: An Efficient

2

9

45

Latent.Space

@latentspacepod

3 months

🆕How to Hire AI Engineers a rare guest post (and bonus pod) from @james_elicit and @_adamwiggins_ ! Covering: - Defining the Hiring Process - Defensive AI Engineering as a chaotic medium - Tech Choices for Defensive AI Engineering - How do you Interview for Defensive AI

1

6

40

Latent.Space

@latentspacepod

11 months

SOTA prompt injection be like:

0

11

41

Latent.Space

@latentspacepod

1 year

🆕 pod on Transformer Math 101! with @QuentinAnthon15 of @AiEleuther everything you ever wanted to know about the nitty gritty of training and making the gpu go brrrrrrrrrrrr

Alessio Fanelli

@FanaHOVA

1 year

The one thing more valuable than GPUs is knowing how to use them (aka make them go brrrr 🔥) @QuentinAnthon15 of @AiEleuther , co-author of "Transformers Math 101", joined @swyx and I on @latentspacepod 🎙️ Take your notepads out before pressing play! We

2

16

50

0

7

35

Latent.Space

@latentspacepod

4 months

🆕 Our first NSFW AI episode, with @itsjessehi ! Covering: - the Waifu vs Fan Platform market - 2-5x'ing the income of OnlyFans creators - Building and Finetuning with DSPy - Model graded evals - Guarding against prompt injections - Memory for

Emulating Humans with NSFW Chatbots - with Jesse Silver

Distilling personality into models with open LLMs, using DSPy and mitigating prompt injection, and 2-5xing the income of OnlyFans creators with the most advanced AI ecommerce chatbot we have ever seen

www.latent.space

7

2

32

Latent.Space

@latentspacepod

1 year

“Cursor is the best product I've used in a while” - @MacCaw “It's so elegant and easy.” - @AndrewMcCalip “Coding with AI is getting insane.” - @MckayWrigley The Latent Space pod is proud to present: the first podcast with @amanrsanger of @anysphere !

Cursor.so: The AI-first Code Editor — with Aman Sanger of Anysphere

On building your own IDE and competing with VSCode and Copilot with a 5 person team, getting in early on GPT-4 + being backed by OpenAI, and why humans should focus on bigger problems than code.

www.latent.space

4

9

33

Latent.Space

@latentspacepod

1 year

🆕 Llama 2: The New Open LLM SOTA >2400 AI Engineers have now tuned into yesterday's recap of Llama 2, ft: - @natolambert of HuggingFace - @BornsteinMatt of a16z - @atroyn of Chroma - @russelljkaplan of Scale and our returning faves @simonw + @altryne !

Llama 2: The New Open LLM SOTA (ft. Nathan Lambert, Matt Bornstein, Anton Troynikov, Russell...

>2000 AI Engineers joined our emergency Space with Simon Willison and Alex Volkov. Discussing datasets, benchmarks, scaling laws, kremlinology, and predictions. Come for the Llama, stay for the drama!

www.latent.space

3

13

31

Latent.Space

@latentspacepod

8 months

OpenAI's RLHF evaluation metrics show us that RLHF can improve some metrics, but it doesn't always mean people are going to be happy with the changes - with @natolambert

2

11

23

Latent.Space

@latentspacepod

9 months

🆕 The AI-First Graphics Editor We talked with @suhail on: - his journey from Mixpanel to Mighty - How a conversation with @model_mechanic inspired @playground_ai - raising $40m to reimagine the graphics editor - how to self-teach AI with

The AI-First Graphics Editor - with Suhail Doshi of Playground AI

Raising $40m to reimagine the graphics editor with AI, self-teaching from no AI background to training (from scratch) a fully open source image generation model that beast SDXL

www.latent.space

4

7

25

Latent.Space

@latentspacepod

15 days

i should call her

fofr

@fofrAI

15 days

13

11

219

1

2

27

Latent.Space

@latentspacepod

2 months

some AWESOME presentations at the @latentspacepod Survey Paper Club today: - @honicky 's guide to ColBERT ( @lateinteraction ): - @dimfeld 's guide to The Landscape of Emerging AI Agent Architectures ( @alexchaomander ): -

0

5

26

Latent.Space

@latentspacepod

8 months

🆕 NeurIPS 2023 Recap — Top Startups! part 2 of our NeurIPS coverage is the biggest ensemble of founders and company leaders we’ve ever featured: • @jefrankle - Chief Scientist, MosaicML • @lqiao - CEO, Fireworks AI • @amanrsanger - CEO, Anysphere

NeurIPS 2023 Recap - AI Startups

Listen now | Mosaic/Databricks, Fireworks, Cursor, Perplexity, Metaphor, Answer.ai, Cerebras, Voxel51, Lightning, Cohere

www.latent.space

1

12

24

Latent.Space

@latentspacepod

8 months

🆕 pod: RLHF 201 Our deep dive into Reinforcement Learning with Human Feedback, with @natolambert of @allen_ai + @interconnectsai ! Covering: - History of RL and its applications - Intellectual history of RLHF dating to Aristotle (!) - RLHF for

1

6

23

Latent.Space

@latentspacepod

1 month

🚨 EMERGENCY PAPER CLUB The @latentspacepod discord is meeting in 2hrs to talk thru @lvdmaaten et al's The Llama 3 Herd of Models, early contender to win the POTY* Awards! Join us (link below) with @swyx , @vibhuuuus , @picocreator , @eugeneyan , et al! *Paper of The Year, totally

2

6

23

Latent.Space

@latentspacepod

7 months

defining inner loop (in a git commit) vs outer loop (after the commit) AI agents. Do you think we'll have true outer loop agents? if so, when?

5

6

22

Latent.Space

@latentspacepod

6 days

🆕 Why you should write your own LLM benchmarks w/ Nicholas Carlini of @GoogleDeepMind Covering his greatest hits: - How I Use AI - My benchmark for large language models - Extracting Training Data from Large Language Models (RIP @openai logprobs) Full episode below!

2

10

23

Latent.Space

@latentspacepod

1 year

FYI: we are holding a snap @latentspacepod on Llama 2 at 4pm PT today! ft. @natolambert @simonw @BornsteinMatt and ofc @fanahova and @swyx !

1

5

21

Latent.Space

@latentspacepod

6 months

Congrats @xai and @elonmusk on the open source release 🫡 Grok-1 cheat sheet 📝 - MoE architecture for 314B total params - 8 experts with 2 active - Apache 2.0 license - Trained on JAX and Rust - Finished training October 2023 - Base model, no task-specific fine-tune - No

1

2

21

Latent.Space

@latentspacepod

1 year

Thanks to @eugeneyan for starting the LS Discord #paperclub and @youngphlo_ our fancy new logo! If you want a welcoming, unrecorded space to dive in to LLM papers every week, we're still open to new members on the LS Discord/Paper club. join in!

2

5

23

Latent.Space

@latentspacepod

2 months

🆕 State of the Art: Training >70B LLMs We are excited to deep dive into @imbue_ai 's incredible new releases with @joshalbrecht (CTO of Imbue) AND with the best return GUEST COHOST possible for the job: @jefrankle (Chief AI Scientist of @DbrxMosaicAI

Imbue

@imbue_ai

2 months

Early this year, we trained a 70B model optimized for reasoning and coding. This model roughly matches LLAMA 3 70B despite being trained on 7x less data. Today, we’re releasing a toolkit to help others do the same, including: • 11 sanitized and extended NLP reasoning

6

83

460

1

5

20

Latent.Space

@latentspacepod

6 months

the Latent Space podcast turns 1 today! reflections from @fanahova (with @swyx ):

Alessio Fanelli

@FanaHOVA

7 months

One Year of Latent Space 🔭 Friday is @latentspacepod birthday, so I wrote out some thoughts on our first year of hosting it with @swyx . Grateful for all the support, and a special s/o to @OfficialLoganK for helping us kick it off as the first guest 🚀

13

102

1

4

20

Latent.Space

@latentspacepod

2 months

🆕 pod alert: Special @aidotengineer Preview Edition! We're so excited to flip the tables with @RazRazcle and help him launch his new podcast, High Agency. A full interview on the 1 year anniversary of the Rise of the AI Engineer, and special preview of

[High Agency] AI Engineer World's Fair Preview

The AIE World's Fair starts tomorrow! A special guest crosspost with Raza Habib covering the 1 year anniversary of the Rise of the AI Engineer, and previewing the Fair.

www.latent.space

0

5

19

Latent.Space

@latentspacepod

4 months

@swyx in this house we stan

[AINews] Lilian Weng on Video Diffusion

AI News for 4/15/2024-4/16/2024. We checked 5 subreddits and 364 Twitters and 27 Discords (395 channels, and 5610 messages) for you. Estimated reading time...

buttondown.com

0

17

Latent.Space

@latentspacepod

5 months

Q: How can 1-bit LLMs match 16-bit LLaMAs? 2-3x faster inference and 20-40x more energy efficient and slightly better evals!! Great work by RJ for the @latentspacepod paper club - a deep dive into the 1-bit LLMs paper! We had a LOT of fun going thru the details this week and

RJ Honicky

@honicky

5 months

Please check out my first step of my in-public education about deep learning Are All Large Language Models Really in 1.58 Bits? { by rj45 }

3

6

53

0

5

18

Latent.Space

@latentspacepod

1 year

🆕 pod: Heralds of the AI Content Flippening with Youssef Rizk of @wondercraft_ai ! AI content is good and getting better. When will it get SO good that our primary media diet is AI generated?

Heralds of the AI Content Flippening — with Youssef Rizk of Wondercraft.ai

AI has come for our jobs! We ask how a 100% AI podcast broke through the US Top 30, survey the Text to Speech AI market, and ask the hard question: What's his moat as an "API Wrapper" startup?

www.latent.space

2

9

17

Latent.Space

@latentspacepod

7 months

🇸🇬 Singapore AI Engineers! Our second meetup is taking place at @StripeDev HQ on 6 Feb: Join @ivanleomk and @swyx ! For those in other countries, feel free to organize your own and tell us on discord!

AI Engineer Singapore Meetup! · Luma

Come by to hang out with @swyx and other ml folks to talk about all things llms, evals and more. Doesn't matter if you're building a ml startup, deploying…

lu.ma

1

6

16

Latent.Space

@latentspacepod

7 months

🆕 Cloud Intelligence at the speed of 5000 tok/s with @vipulved (CEO) and Ce Zhang (CTO) of @TogetherCompute ! A deep dive into the full stack research lab + GPU cloud behind RedPajama, Mamba, StripedHyena, and why openness is core to Together's

0

7

16

Latent.Space

@latentspacepod

9 months

🆕 We Are Running Out of Low-Background Tokens Was 2022 the last year we had "clean" data uncontaminated by ChatGPT? (h/t @jeffreyhuber for the concept!) also: - @martin_casado 's Voice Cloning Red Alert - that @pika_labs video - incredible LCM demos

2

14

15

Latent.Space

@latentspacepod

6 months

🆕 Weekend pod: Jan+Feb recap + 1 Yr of Latent Space! Our 2023 recap pod was very well received, so here's the next in this new series of research recaps, with takes and guests from: - OpenAI Sora - Gemini Pro 1.5 (shoutout @simonw 's post) - Groq infra

OpenAI Sora, Google Gemini, Groq Math, and our Top 5 Research Trends (Jan-Feb 2024 Audio Recap +...

Listen now | swyx & Alessio discuss the research trends of January and the industry chaos of February 2024. Also: we celebrate the 1 year anniversary of Latent Space!

www.latent.space

0

7

14

Latent.Space

@latentspacepod

11 months

🆕 pod: Building the Foundation Model Ops Platform with @RazRazcle of Humanloop! Covering: Three Stages of Evals & Human Feedback, the market for ops platforms, and diving into the most underrated LLM research. Also asking: Did GPT-4 get dumber?

Building the Foundation Model Ops Platform — with Raza Habib of Humanloop

Finally, our LLMOps episode! We ask the OG LLM/FM/??Ops founder on the Three Stages of Evals & Feedback, the market for ops platforms, and the most underrated LLM research. Plus: Did GPT-4 get dumber?

www.latent.space

0

5

14

Latent.Space

@latentspacepod

7 months

🆕 Why StackOverflow usage is down 50% with @dvdhsu of @Retool ! Chart says it all. More details on Retool AI and Retool's State of AI in today's pod.

Why StackOverflow usage is down 50% — with David Hsu of Retool

Retool's CEO on the State of AI, why Silicon Valley underestimates internal tooling, model adoption in 2023, open source vs OpenAI trade offs, and Retool's journey from YC to a $3.2B company

www.latent.space

1

4

13

Latent.Space

@latentspacepod

5 months

New weekend pod alert! this is a long 'un, something in it for everyone, whether you're into @Grok or @GroqInc , or all about Corporate AI Leadership or Personal AI therapists. Enjoy and please send feedback!

swyx.io

@swyx

5 months

and our meetup recording is live, together with 4 other podcast appearances with @nlw , @RohdeAli / @joshalbrecht , @SemiAnalysis_ , et al, on this weekend's @latentspacepod drop!

3

17

1

3

12

Latent.Space

@latentspacepod

1 year

Recording for Saturday's LLM Finetuning event is here! Grateful to @LangChainAI @brevdev @replicatehq @anyscalecompute @metaphorsystems @_FireworksAI @phindsearch and everyone who came and hacked! (recording turned out bad unfortunately so unlisted fn)

Anything but Wrappers: Finetuning ft. Langchain, Anyscale, Phind,...

Due to overwhelming interest in finetuning, (https://twitter.com/AlexReibman/status/1695643709889818969?s=20) we put on a one day event and invited a panel ...

www.youtube.com

Alex Reibman 🖇️

@AlexReibman

1 year

LLM Fine tuning is here! San Francisco’s top AI engineers came together to see what’s possible with fine-turning and only 4 hours of hacking. Here’s an exclusive what we saw at the “Anything But Wrappers” hackathon (🧵):

54

326

3K

1

4

12

Latent.Space

@latentspacepod

3 months

@james_elicit @_adamwiggins_ Full post and shownotes:

How To Hire AI Engineers — with James Brady & Adam Wiggins of Elicit

On Defensive vs Offensive AI Engineering and the ML First mindset: Presenting our ultimate guide to Hiring AI Engineers (and How to Source Them)!

www.latent.space

0

2

12

Latent.Space

@latentspacepod

1 month

🆕 pod: The Winds of AI Winter! The vibes have shifted... @fanahova and @swyx celebrate 1m downloads and recap the last 3 months in AI! Discussing the Frontier Labs vibe shift between Claude 3.5, Llama 3.1, Apple Intelligence, and the expansion of

1

11

Latent.Space

@latentspacepod

2 months

New pod live with @clefourrier on the Open LLM Leaderboard V2 and the state of LLM Evaluation!

swyx.io

@swyx

2 months

🆕Benchmarks 201 with @clefourrier of the @huggingface Open LLM Leaderboard! Why Chat Arenas are **OVERRATED**: - they bias sycophancy/assertiveness over factuality - annotators are not representative of humanity - models are argmaxxing public arena

9

21

67

0

1

11

Latent.Space

@latentspacepod

2 months

@elonmusk @Tesla @xai Can we do a live X Space from the data center when the first workload goes live? 🔌

1

0

9

Latent.Space

@latentspacepod

2 months

CALLING ALL @LATENTSPACEPOD FANS Join our @aidotengineer IRL meetup!! Tonight 7.30pm in SF

Latent Space Discord meetup · Luma

Lots of folk from Discord flew in for the AIEWF conference. Let's meet up! Tropisueno has good food, fantastic drinks, and if it's all packed there are plenty…

lu.ma

2

1

11

Latent.Space

@latentspacepod

10 months

NYC 🗽 our meetup tmr is confirmed! ~40 AI folks coming. come by! singing optional!

Latent Space // Devtools NYC drinks (SUNDAY 5pm) | Partiful

Latent Space/Devtools discord and friends hang out! drinks on own dime mostly. but also come sing!

partiful.com

0

1

11

Latent.Space

@latentspacepod

10 months

> be @realGeorgeHotz > jailbreak 1st iPhone > trade that for Nissan 350Z > jailbreak PS3 > get sued by Sony > write a rap about it

1

10

Latent.Space

@latentspacepod

2 months

proudly fine

Grant♟️

@granawkins

2 months

@dylan522p @NoPriorsPod @dwarkesh_sp @latentspacepod Latent space is fine. Dwarkesh is not an AI podcast. Both of the above regularly have long convos with engineers about what they’re doing today that is sota.

1

0

1

0

9

Latent.Space

@latentspacepod

10 months

We asked @realGeorgeHotz what the AI enabled future looks like? 2 words 👉 AI Girlfriend

4

3

9

Latent.Space

@latentspacepod

8 months

🔜

1

9

Latent.Space

@latentspacepod

21 days

thanks to @honicky and @vibhuuuus for another great Paper Club on ReFT: Representation Fine Tuning by @aryaman2020 et al! signups below

1

9

Latent.Space

@latentspacepod

8 months

thanks Leila! yes the philosophy is to just have an ongoing conversation with listeners and honestly a dumping ground for all the stuff that we’re not yet sure about but found noteworthy anyway the quality of the @eugeneyan @yikesawjeez @YoungPhlo_ @eugenechh et al paper club

Leila Clark

@leilavclark

8 months

@dwarkesh_sp I love the *idea* of a subscriber-only discord, but I haven’t yet seen it done well. In general people who post a lot in a large discords usually aren’t very interesting. The only exception I know is the @latentspacepod discord where swyx/Alessio post a lot of good stuff.

0

6

1

4

8

Latent.Space

@latentspacepod

8 months

Huge congrats to Perplexity! first and most well deserved AI raise of 2024. check out our NeurIPS convo with Aravind here: any bets on where Perplexity will end the year?

NeurIPS 2023 Recap - AI Startups

Listen now | Mosaic/Databricks, Fireworks, Cursor, Perplexity, Metaphor, Answer.ai, Cerebras, Voxel51, Lightning, Cohere

www.latent.space

Aravind Srinivas

@AravSrinivas

8 months

Excited to announce we've raised 73.6M$ at 520M$ valuation, led by IVP, along with our seed and Series A lead investors NEA, Elad Gil, Nat Friedman. Honored to have NVIDIA, Jeff Bezos, @tobi , Databricks, @naval , @rauchg , @balajis partner in this round, among others.

418

292

6K

0

1

8

Latent.Space

@latentspacepod

6 months

a bit delayed, but @bernhardsson 's old post is frontpaging on HN today. If you have opinions on what the ideal AI infra platform is, chime in!

Alessio Fanelli

@FanaHOVA

7 months

🆕 Truly Serverless Infra for AI Engineers In 2021 @bernhardsson wrote a "Software Infra 2.0 wishlist", but quickly decided to take matter into his own hands by building @modal_labs . On today's @latentspacepod we explain how VM-based clouds don't work

2

15

60

0

1

8

Latent.Space

@latentspacepod

8 months

there can be some interesting (but also hilarious) quirks when generating functional UI on @tldraw as we found out with @steveruizok 👀

1

4

8

Latent.Space

@latentspacepod

10 months

how do you get @paulg to personally invest in you?

0

5

8

Latent.Space

@latentspacepod

5 months

@jxnlco we work hard for the RT boss

0

7

Latent.Space

@latentspacepod

10 months

. @MichaelRoyzen literally made a local vision model That could run on a phone With no internet connection That was better than Google Lens In 2017!

0

5

7

Latent.Space

@latentspacepod

7 months

h/t @picocreator in latent space paper club today

1

7

Latent.Space

@latentspacepod

28 days

1

0

7

Latent.Space

@latentspacepod

1 year

experimenting with a "monthly recap newsletter" - let us know if you like this!

swyx.io

@swyx

1 year

🆕 AI Eng Recap: August 2023 My compilation of top themes and news for AI Engineers. Enjoy!

4

23

118

0

1

7

Latent.Space

@latentspacepod

2 months

@aiDotEngineer huge turnout!

0

1

7

Latent.Space

@latentspacepod

9 months

@JeffDean @rm_rafailov @Muennighoff @Tim_Dettmers @sy_gadre @imhaotian ahh too late for us to change but acknowledged! figured he wasnt active on twitter so tagged you 🙇‍♀️

0

7

Latent.Space

@latentspacepod

4 months

At Stitch Fix, @jxnlco created a $50MM money printer. They built a recommendation engine using embeddings of the comments, reviews, and clothing. This let them recommend items that work in the same outfit.

1

7

Latent.Space

@latentspacepod

15 days

This is entirely speculative but... Tomorrow's LS paper club with @picocreator is going to be extremely lit! come learn about the state of the art in Speculative Decoding!

Latent.Space

@latentspacepod

21 days

thanks to @honicky and @vibhuuuus for another great Paper Club on ReFT: Representation Fine Tuning by @aryaman2020 et al! signups below

1

9

3

2

8

Latent.Space

@latentspacepod

8 months

What if there are more modalities than meets the eye? What could be built out of that? What problems could be solved? - @Suhail with a potential moonshot company and going beyond current modalities with physics

0

3

6

Latent.Space

@latentspacepod

8 months

@chr1sa @CogRev_Podcast @NoPriorsPod wow thanks Chris! that is high praise, we look up to @labenz and @saranormous and @eladgil in our work 🙇‍♂️

1

0

5

Latent.Space

@latentspacepod

8 months

our quotes page just be growing tbh pls keep em coming! really helps us book guests!!

Chris Anderson

@chr1sa

8 months

I listen to a ton of AI podcasts and by far the best are where the host is a technical expert rather than a journalist or other generalist. Deeper, smarter, more in tune with their audience The gold standards are @latentspacepod , @CogRev_Podcast and @NoPriorsPod IMO

12

16

133

0

6

5

Latent.Space

@latentspacepod

19 days

our next #ai -in-action club is starting now: on @lateinteraction ’s DSPy! led by @ProgramWithAi and @kbal11

2

5

Latent.Space

@latentspacepod

10 months

"Pay attention to that crazy demo you can't forget about ... It's much easier to stay obsessed every single day" - @MichaelRoyzen

0

5

Latent.Space

@latentspacepod

1 month

The Winds of AI Winter

Mar-Jun 2024 Recap: People are raising doubts about AI Summer. Here's why AI Engineers are the solution.

www.latent.space

0

5

Latent.Space

@latentspacepod

5 months

NYC @latentspacepod friends: @VikParuchuri is running a smol AI meetup in Brooklyn this Friday! join us!

Smol AI meetup · Luma

We're doing an AI meetup - join to discuss anything AI-related, but especially model architectures, recent interesting papers, and training.

lu.ma

0

3

5

Latent.Space

@latentspacepod

10 months

new pod dropped today for the GPU Poors!

swyx.io

@swyx

10 months

🆕 We got that semianalysis guy @dylan522p on @latentspacepod ! Dishing on: - what it's like to live the GPU Rich/Poor divide - Why GPT4 cost $500m but training costs are irrelevant - non-Nvidia silicon (AMD/TPU/Cerebras/Matx) - Can we ever move semis

5

14

118

0

1

5

Latent.Space

@latentspacepod

6 months

@AstasiaMyers @hhua_ @FanaHOVA @dylan522p @swyx had to manage the q&a polls while sharing my screen for said q&a haha

0

2

Latent.Space

@latentspacepod

2 months

@imbue_ai @joshalbrecht @jefrankle @DbrxMosaicAI join us on youtube for full discussion!

State of the Art: Training 70B LLMs on 10,000 H100 clusters

Josh Albrecht, CTO of Imbue, and Jon Frankle, Chief AI Scientist of Databricks, dish on what it takes to train the largest models on the largest clusters... ...

www.youtube.com

0

1

4

Latent.Space

@latentspacepod

5 months

"I want AI to know me really, really well" - Igor His wild approach? Recording every single conversation to build a massive personal audio dataset 1 year of his life's audio is less than 1 terabyte of data

3

2

4

Latent.Space

@latentspacepod

3 months

WWDC watch party going on in LS Discord right now! (search Latent Space Discord to join)

0

4

Latent.Space

@latentspacepod

2 months

@jimmy6DOF @YiTayML @RekaAILabs @sama @_jasonwei tbh i think i added one too many images haha

0

4

Latent.Space

@latentspacepod

5 months

ICYMI: our new pod with @elicitorg is out! Lots of insights on the journey to build the ultimate AI Research Assistant

Supervise the Process of AI Research — with Jungwon Byun and Andreas...

Timestamps:00:00:00 Introductions00:07:45 How Johan and Andreas Joined Forces to Create Elicit00:10:26 Why Products are better than Research00:15:49 The Evol...

www.youtube.com

swyx.io

@swyx

5 months

Cost v Reliability v Performance: Choose Three! 🆕 pod: w/ @jungofthewon and @stuhlmueller How do you improve prompt engineering and ai agents? The emerging paradigm is sometimes called "Flow engineering", but @elicitorg has reinvented the Notebook, in

1

9

84

0

6

3

Latent.Space

@latentspacepod

3 months

@dpkingma forgot the link!

ICLR 2024 — Best Papers & Talks (ImageGen, Vision, Transformers, State Space Models) ft. Durk...

14 of the best papers out of the 2260 papers presented at the 2024 ICLR conference, in 4 sections covering Image Generation, Vision Learning, Extending Transformers, and State Space Models.

www.latent.space

1

4

Latent.Space

@latentspacepod

2 months

@fleetwood___ at what quantization? dont forget that they do 3.5 bit

1

0

4

Latent.Space

@latentspacepod

1 year

just a week after we talked about it, @OpenAI launched voice synthesis and dubbing on Spotify!

Greg Brockman

@gdb

1 year

Speak in your voice in the listener’s language, not your own, using the latest @OpenAI voice synthesis technology. Great use-case by Spotify:

36

159

1K

1

3

4

Latent.Space

@latentspacepod

8 months

SF Meetup announcement: @Swizec is organizing a casual meetup this Friday at SPIN! join us to talk AI, dev tools, and whack away at ping pong

Latent Space SF Meetup · Luma

Casual meetup! not huge, just social. Table booking is for 6.30pm, but some of us will be there from 6pm. Check Latent Space Discord for details and live chat.

lu.ma

1

9

4

Latent.Space

@latentspacepod

29 days

@peterchng @hu_yifei @Xianbao_QIAN @Senseye_Winning ah yes our classic from @QuentinAnthon15 !

Latent.Space

@latentspacepod

1 year

🆕 pod on Transformer Math 101! with @QuentinAnthon15 of @AiEleuther everything you ever wanted to know about the nitty gritty of training and making the gpu go brrrrrrrrrrrr

0

7

35

1

0

4

Latent.Space

@latentspacepod

1 year

guess the next latent space pod guest 👀

Aman Sanger

@amanrsanger

1 year

Despite Cursor’s recent insane growth, the current version is just 0.1% of what we have in store. We’re a small, very strong team and are looking for fantastic SWEs and designers to help shape the future of software development. Read more here -

19

18

354

0

3

4

Latent.Space

@latentspacepod

2 months

And we somehow forgot to post ICLR pod part 2! First: a good ol' longform convo with @gneubig on his work, ft. OpenDevin which has now formed into a company! and our first-ever returning-guest guest-host, @amanrsanger !

Graham Neubig

@gneubig

2 months

Announcement: @rbren_dev , @xingyaow_ , and I have formed a company! Our name is All Hands AI 🙌 And our mission is to build the world’s best AI software development agents, for everyone, in the open. Here’s why I think this mission is important 🧵

32

97

720

1

4

Latent.Space

@latentspacepod

8 days

Guess the guest!

Alessio Fanelli

@FanaHOVA

8 days

Midjourney web generations (right) feel extremely different than Discord ones (left). It's extremely biased towards making cartoonish / drawing images. Haven't tweaked any setting. Same prompt, ran 2-3 times each; all web ones were drawings, ~20% of Discord ones were. "a

3

0

4

0

3

4

Latent.Space

@latentspacepod

8 months

@natolambert episode ref w/ timestamp:

The Origin and Future of RLHF: the secret ingredient for ChatGPT -...

The origins of Reinforcement Learning from Human Feedback, RLHF, sociology's influence on it, the tension between human vs synthetic data, and emerging resea...

www.youtube.com

0

1

3

Latent.Space

@latentspacepod

6 months

New pod live! for audio and full show notes!

Open Source AI is AI we can Trust — with Soumith Chintala of Meta AI

Listen now | The PyTorch creator riffs on geohot's Tinygrad, Chris Lattner's Mojo, Apple's MLX, the PyTorch Mafia, the upcoming Llama 3 and MTIA ASIC, AI robotics, and what it takes for open source...

www.latent.space

swyx.io

@swyx

6 months

🆕 The @latentspacepod is excited to present: Open Source AI with @soumithchintala ! Covering: - Meta's 600k H100-equivalent GPU compute - PyTorch vs @realGeorgeHotz 's tinygrad vs @clattner_llvm 's Mojo vs Apple MLX - PyTorch Mafia and the Mixtral

5

20

180

0

2

3

Latent.Space

@latentspacepod

10 months

signal boosting this important critique of the FMTI BY @natolambert , @BlancheMinerva , @aviskowron et al: in retrospect any transparency rating that rates gpt4 the third most transparent FM in the world, above @EMostaque ’s StableLM, is kiiinda sus

How the Foundation Model Transparency Index Distorts Transparency

A proper critique of the Foundation Model Transparency Index (FMTI). Plus some thoughts on the ecosystem implications.

www.interconnects.ai

EleutherAI

@AiEleuther

10 months

The Foundation Model Transparency Index by @StanfordCRFM purports to be an assessment of how transparent popular AI models are. Unfortunately its analysis is quite flawed in ways that minimize its usefulness and encourage gamification

7

35

150

0

5

3