sankalp (dejavucoder)
@dejavucoder.bsky.social
into applied ai + product engg
interested in all things ai and distributed systems
interested in all things ai and distributed systems
bookmarking here to read this soon
alexzhang13.github.io/blog/2024/ef...
alexzhang13.github.io/blog/2024/ef...
Alex L. Zhang | A Meticulous Guide to Advances in Deep Learning Efficiency over the Years
A very long and thorough guide how deep learning algorithms, hardware, libraries, compilers, and more have become more efficient.
alexzhang13.github.io
January 9, 2025 at 8:21 PM
bookmarking here to read this soon
alexzhang13.github.io/blog/2024/ef...
alexzhang13.github.io/blog/2024/ef...
Reposted by sankalp (dejavucoder)
The state of post-training in 2025: a tutorial on modern post-training
A re-record of my NeurIPS tutorial on language modeling (plus some added content on the high level state of things)
Blog + extra context: https://buff.ly/424VvLm
YouTube: https://buff.ly/40808l5
Slides: https://buff.ly/404jGa9
A re-record of my NeurIPS tutorial on language modeling (plus some added content on the high level state of things)
Blog + extra context: https://buff.ly/424VvLm
YouTube: https://buff.ly/40808l5
Slides: https://buff.ly/404jGa9
The state of post-training in 2025
A re-record of my NeurIPS tutorial on language modeling (plus some added content).
buff.ly
January 8, 2025 at 3:38 PM
The state of post-training in 2025: a tutorial on modern post-training
A re-record of my NeurIPS tutorial on language modeling (plus some added content on the high level state of things)
Blog + extra context: https://buff.ly/424VvLm
YouTube: https://buff.ly/40808l5
Slides: https://buff.ly/404jGa9
A re-record of my NeurIPS tutorial on language modeling (plus some added content on the high level state of things)
Blog + extra context: https://buff.ly/424VvLm
YouTube: https://buff.ly/40808l5
Slides: https://buff.ly/404jGa9
new blog post
Evolution of AI-assited coding features and developer interaction patterns. I go through the history of progression of ai-assisted coding features, talk about how we interact with them and a Gears analogy control vs speed tradeoff
sankalp.bearblog.dev/evolution-of...
Evolution of AI-assited coding features and developer interaction patterns. I go through the history of progression of ai-assisted coding features, talk about how we interact with them and a Gears analogy control vs speed tradeoff
sankalp.bearblog.dev/evolution-of...
The Evolution of AI-assisted coding features and developer interaction patterns
Yes, I agree that's a fancy title. There have been several developments over the last 7 years in the AI-assisted coding arena. We have gone from simple autoc...
sankalp.bearblog.dev
December 21, 2024 at 7:54 PM
new blog post
Evolution of AI-assited coding features and developer interaction patterns. I go through the history of progression of ai-assisted coding features, talk about how we interact with them and a Gears analogy control vs speed tradeoff
sankalp.bearblog.dev/evolution-of...
Evolution of AI-assited coding features and developer interaction patterns. I go through the history of progression of ai-assisted coding features, talk about how we interact with them and a Gears analogy control vs speed tradeoff
sankalp.bearblog.dev/evolution-of...
Reposted by sankalp (dejavucoder)
First slide deck for NeurIPS is done -- an overview of how I view post-training for applications.
A higher level summary on the key decisions along the way of scoping a problem, choosing a base model, optimization algorithm, etc. (+some thoughts on OpenAI's RL Finetuning).
https://buff.ly/3ZpY5IR
A higher level summary on the key decisions along the way of scoping a problem, choosing a base model, optimization algorithm, etc. (+some thoughts on OpenAI's RL Finetuning).
https://buff.ly/3ZpY5IR
December 9, 2024 at 7:04 PM
First slide deck for NeurIPS is done -- an overview of how I view post-training for applications.
A higher level summary on the key decisions along the way of scoping a problem, choosing a base model, optimization algorithm, etc. (+some thoughts on OpenAI's RL Finetuning).
https://buff.ly/3ZpY5IR
A higher level summary on the key decisions along the way of scoping a problem, choosing a base model, optimization algorithm, etc. (+some thoughts on OpenAI's RL Finetuning).
https://buff.ly/3ZpY5IR
agent orchestrator more like agent pimp
December 4, 2024 at 5:09 PM
agent orchestrator more like agent pimp
will check this out for synthetic data creation and evals
smol course Day 1 ✅. I learnt that people are hungry for models they can own.
📚 Material focused on instruction tuning. Split into chat templates and supervised fine tuning. There's more to this subject than this, but we're keeping things smol.
⏩ If you haven't already, try out module 1!
🧵
📚 Material focused on instruction tuning. Split into chat templates and supervised fine tuning. There's more to this subject than this, but we're keeping things smol.
⏩ If you haven't already, try out module 1!
🧵
December 4, 2024 at 5:08 PM
will check this out for synthetic data creation and evals
Reposted by sankalp (dejavucoder)
New post! OpenAI's o1 using "search" was a PSYOP.
How to understand OpenAI's o1 models as really just one wacky, wonderful, long chain of thought.
A fun one trying to communicate intuitions for what large scale RL training does to LLMs. Much more to explore here in 2025!
How to understand OpenAI's o1 models as really just one wacky, wonderful, long chain of thought.
A fun one trying to communicate intuitions for what large scale RL training does to LLMs. Much more to explore here in 2025!
OpenAI's o1 using "search" was a PSYOP
How to understand OpenAI's o1 models as really just one wacky, wonderful, long chain of thought
buff.ly
December 4, 2024 at 3:33 PM
New post! OpenAI's o1 using "search" was a PSYOP.
How to understand OpenAI's o1 models as really just one wacky, wonderful, long chain of thought.
A fun one trying to communicate intuitions for what large scale RL training does to LLMs. Much more to explore here in 2025!
How to understand OpenAI's o1 models as really just one wacky, wonderful, long chain of thought.
A fun one trying to communicate intuitions for what large scale RL training does to LLMs. Much more to explore here in 2025!
Reposted by sankalp (dejavucoder)
Wow, this is such a useful resource of industry LLM applications! And filtering via search/tags is so responsive. I was thinking of compiling something like this over the holidays (ala applied-ml) but thanks to @strickvl.bsky.social I can spend the time reading instead ♥️
zenml.io/llmops-datab...
zenml.io/llmops-datab...
December 3, 2024 at 1:54 AM
Wow, this is such a useful resource of industry LLM applications! And filtering via search/tags is so responsive. I was thinking of compiling something like this over the holidays (ala applied-ml) but thanks to @strickvl.bsky.social I can spend the time reading instead ♥️
zenml.io/llmops-datab...
zenml.io/llmops-datab...
this is kinda nice
November 26, 2024 at 2:54 PM
this is kinda nice
we are planning to read this blog blog.dottxt.co
November 26, 2024 at 2:43 PM
we are planning to read this blog blog.dottxt.co