@AnthropicAI | prev @uwcse
Joined October 2011
- Tweets 422
- Following 1,054
- Followers 1,900
- Likes 1,366
51 Photos and videos
Pinned Tweet
28 Sep 2023
Sharing some highlights from our work on small-scale proxies for large-scale Transformer training instabilities: arxiv.org/abs/2309.14322
With fantastic collaborators @peterjliu, @Locchiu, @_katieeverett, many others (see final tweet!), @hoonkp, @jmgilmer, @skornblith!
(1/15)
3
61
340
100,508
Mitchell Wortsman retweeted
10 Dec 2025
New job! I’m hiring folks interested in building and researching the next generation of evals and eval infa. DMs are open :)
112
68
2,212
144,890
Mitchell Wortsman retweeted
5 Jun 2025
Very excited to finally release our paper for OpenThoughts!
After DataComp and DCLM, this is the third large open dataset my group has been building in collaboration with the DataComp community. This time, the focus is on post-training, specifically reasoning data.
22
206
1,324
187,522
Mitchell Wortsman retweeted
22 May 2025
Introducing the next generation: Claude Opus 4 and Claude Sonnet 4.
Claude Opus 4 is our most powerful model yet, and the world’s best coding model.
Claude Sonnet 4 is a significant upgrade from its predecessor, delivering superior coding and reasoning.
ALT A benchmarking table titled Claude 4 benchmarks comparing performance metrics across various capabilities including coding, reasoning, tool use, multilingual Q&A, visual reasoning, and mathematics.
928
3,153
20,650
4,286,571
Mitchell Wortsman retweeted
21 May 2025
Excited to share that I'll be joining @Anthropic to work on pretraining science! I've chosen to defer my Stanford PhD, where I'm honored to be supported by the Hertz Fellowship.
There's something special about the science, this place, and these people. Looking forward to joining some of my most brilliant and compassionate colleagues!
42
10
755
58,821
Mitchell Wortsman retweeted
19 May 2025
Many agents (Claude Code, Codex CLI) interact with the terminal to do valuable tasks, but do they currently work well enough to deploy en masse?
We’re excited to introduce Terminal-Bench: An evaluation environment and benchmark for AI agents on real-world terminal tasks. Tl;dr lots of room for improvement! tbench.ai/
16
63
243
52,219
Mitchell Wortsman retweeted
26 Apr 2025
Excited to be presenting at #ICLR2025 at 10am today on how generative classifiers are much more robust to distribution shift. Come by to chat and say hello!
2
5
90
6,559
Mitchell Wortsman retweeted
12 Dec 2024
I'm presenting our #NeurIPS2024 work on Attention Transfer today!
Key finding: Pretrained representations aren't essential - just using attention patterns from pretrained models to guide token interactions is enough for models to learn high-quality features from scratch and match ImageNet performance! 🤯
Chat with me and @endernewton Dec 12 (today), 4:30 -7:30 pm PST, East Exhibit Hall #1900
2
21
155
14,125
Mitchell Wortsman retweeted
4 Dec 2024
🚨 I’m on the job market this year! 🚨
I’m completing my @uwcse Ph.D. (2025), where I identify and tackle key LLM limitations like hallucinations by developing new models—Retrieval-Augmented LMs—to build more reliable real-world AI systems. Learn more in the thread! 🧵
ALT Overview of Akari's research. More information is at https://akariasai.github.io/
27
117
811
126,777
Mitchell Wortsman retweeted
4 Dec 2024
I'm on the academic job market!
I develop autonomous systems for: programming, research-level question answering, finding sec vulnerabilities & other useful challenging tasks.
I do this by building frontier-pushing benchmarks and agents that do well on them.
See you at NeurIPS!
9
39
230
24,007
Mitchell Wortsman retweeted
22 Oct 2024
Introducing an upgraded Claude 3.5 Sonnet, and a new model, Claude 3.5 Haiku. We’re also introducing a new capability in beta: computer use.
Developers can now direct Claude to use computers the way people do—by looking at a screen, moving a cursor, clicking, and typing text.
ALT A benchmark comparison table showing performance metrics for multiple AI models including Claude 3.5 Sonnet (new), Claude 3.5 Haiku, GPT-4o, and Gemini models across different tasks.
467
1,826
9,989
3,698,983
Mitchell Wortsman retweeted
17 Oct 2024
OpenCLIP passed 10K stars on GitHub this week. A big milestone for any open-source project. 🍻 to the many collaborators that made that possible. Coincidentally, I pushed a new release with a port of the largest multi-lingual SigLIP -- a SO400M/16 @ 256x256 that appeared on big_vision a little while back.
Now on the @huggingface hub and useable via timm or OpenCLIP (update your timm too)! huggingface.co/timm/ViT-SO40…
5
15
147
26,228
Mitchell Wortsman retweeted
23 Jul 2024
Come chat with me and @Locchiu at our ICML poster session 1:30-3pm CEST (Vienna time) today at Hall C 4-9 #2500 and see how our theory lets all parameterizations perform hyperparameter transfer!
arxiv.org/abs/2407.05872
6
3
28
65,737
Mitchell Wortsman retweeted
18 Jul 2024
We have released our DCLM models on huggingface! To our knowledge these are by far the best performing truly open-source models (open data, open weight models, open training code) 1/5
9
62
285
51,222
Mitchell Wortsman retweeted
18 Jul 2024
We've gotten some great questions about the notion of alignment in our width-scaling parameterization paper! arxiv.org/abs/2407.05872
A deep dive into the alignment metric and intuition 🧵 [1/16]
4
17
66
14,818
Mitchell Wortsman retweeted
2 Jul 2024
🧵1/8 We resolve the discrepancy between the compute optimal scaling laws of Kaplan (exponent 0.88, Figure 14, left) et al. and Hoffmann et al. (“Chinchilla”, exponent 0.5).
Paper: arxiv.org/abs/2406.19146
Data Code: github.com/formll/resolving-…
6
34
170
36,040
Mitchell Wortsman retweeted
20 Jun 2024
We're also launching a preview of Artifacts on claude.ai.
You can ask Claude to generate docs, code, mermaid diagrams, vector graphics, or even simple games.
Artifacts appear next to your chat, letting you see, iterate, and build on your creations in real-time.
37
188
1,630
588,443
Mitchell Wortsman retweeted
20 Jun 2024
Introducing Claude 3.5 Sonnet—our most intelligent model yet.
This is the first release in our 3.5 model family.
Sonnet now outperforms competitor models on key evaluations, at twice the speed of Claude 3 Opus and one-fifth the cost.
Try it for free: claude.ai
ALT Benchmark table showing Claude 3.5 Sonnet outperforming (as indicated by green highlights) other AI models on graduate level reasoning, code, multilingual math, reasoning over text, and more evaluations. Models compared include Claude 3 Opus, GPT-4o, Gemini 1.5 Pro, and Llama-400b.
419
1,531
7,040
2,522,663
Mitchell Wortsman retweeted
19 Jun 2024
Thrilled to share our paper “Large-Scale Transfer Learning for Tabular Data via Language Modeling,” introducing TabuLa-8B: a foundation model for prediction on tabular data.
(with Juan C Perdomo @lschmidt3)
📖 arxiv.org/abs/2406.12031
🌐 huggingface.co/collections/m…
[long🧵]
3
4
23
5,335
Mitchell Wortsman retweeted
18 Jun 2024
I am really excited to introduce DataComp for Language Models (DCLM), our new testbed for controlled dataset experiments aimed at improving language models. 1/x
7
79
273
120,142
Mitchell Wortsman retweeted
5 Jun 2024
We recently open-sourced a relatively minimal implementation example of Transformer language model training in JAX, called NanoDO.
If you stick to vanilla JAX components, the code is relatively straightforward to read -- the model file is <150 lines. We found it useful as a fork-able example for researchers to easily hack on and experiment rapidly. While we do not provide SOTA configs, we hope it can help researchers get started to try ideas, especially if new to LMs and JAX -- which is truly under-rated.
Repo: github.com/google-deepmind/n…
3
58
278
58,921