@DanielaAmodei profile picture
Daniela Amodei @DanielaAmodei

President @AnthropicAI. Formerly @OpenAI, @Stripe, congressional staffer, global development

Joined September 2011
  • Posts 29
  • Following 289
  • Followers 29,673
1 Photos and videos
Daniela Amodei reposted
Anthropic
 @AnthropicAI
11 Jul 2023
Introducing Claude 2! Our latest model has improved performance in coding, math and reasoning. It can produce longer responses, and is available in a new public-facing beta website at claude.ai in the US and UK.
262
493
2,298
863,589
Daniela Amodei reposted
Anthropic
 @AnthropicAI
14 Sep 2022
Neural networks often pack many unrelated concepts into a single neuron – a puzzling phenomenon known as 'polysemanticity' which makes interpretability much more challenging. In our latest work, we build toy models where the origins of polysemanticity can be fully understood.
59
627
3,875
Daniela Amodei reposted
Anthropic
 @AnthropicAI
13 Jul 2022
In "Language Models (Mostly) Know What They Know", we show that language models can evaluate whether what they say is true, and predict ahead of time whether they'll be able to answer questions correctly. arxiv.org/abs/2207.05221
20
154
928
Daniela Amodei reposted
Anthropic
 @AnthropicAI
27 Jun 2022
Transformer MLP neurons are challenging to understand. We find that using a different activation function (Softmax Linear Units or SoLU) increases the fraction of neurons that appear to respond to understandable features without any performance penalty. transformer-circuits.pub/202…
11
71
384
Daniela Amodei reposted
Anthropic
 @AnthropicAI
24 May 2022
In a new paper, we show that repeating only a small fraction of the data used to train a language model (albeit many times) can damage performance significantly, and we observe a "double descent" phenomenon associated with this. arxiv.org/abs/2205.10487
9
41
335
I’m looking forward to what’s to come. And we’re hiring! anthropic.com/#careers

Home \ Anthropic

Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

anthropic.com
8
2
33
As well as steerability and robustness -arxiv.org/abs/2112.00861 - reinforcement learning - arxiv.org/abs/2204.05862, societal impacts - arxiv.org/abs/2202.07785, and more!

A General Language Assistant as a Laboratory for Alignment

Given the broad capabilities of large language models, it should be possible to work towards a general-purpose, text-based assistant that is aligned with human values, meaning that it is helpful,...

arxiv.org
3
2
24
This includes work on interpretability - transformer-circuits.pub/202… and transformer-circuits.pub/202… - and some interpretability resources like Garcon transformer-circuits.pub/202… and Pysvelte github.com/anthropics/PySvel…
1
6
I’m so proud of the amazing team we’ve assembled at Anthropic and the research we’ve done to date.
1
9
Excited to announce our latest fundraising round! We’re genuinely honored to be entrusted with the resources to continue our work in frontier AI safety and research.
Anthropic
 @AnthropicAI
29 Apr 2022
We’ve raised $580 million in a Series B. This will help us further develop our research to build usable, reliable AI systems. Find out more: anthropic.com/news/announcem…
25
9
106
Daniela Amodei reposted
Anthropic
 @AnthropicAI
15 Apr 2022
Glad @QuantaMagazine highlights progress on induction heads/rigorous interpretability by @ch402, @catherineols, @nelhage and others @AnthropicAI. More to come! quantamagazine.org/researche…

Researchers Glimpse How AI Gets So Good at Language Processing | Quanta Magazine

Language processing programs are notoriously hard to interpret, but smaller versions can provide important insights into how they work.

quantamagazine.org
2
11
78
Daniela Amodei reposted
Anthropic
 @AnthropicAI
13 Apr 2022
We've trained a natural language assistant to be more helpful and harmless by using reinforcement learning with human feedback (RLHF). arxiv.org/abs/2204.05862
3
48
267
Daniela Amodei reposted
Anthropic
 @AnthropicAI
9 Mar 2022
On the @FLIxrisk podcast, we discuss AI research, AI safety, and what it was like starting Anthropic during COVID. futureoflife.org/2022/03/04/…
5
10
53
Daniela Amodei reposted
Anthropic
 @AnthropicAI
8 Mar 2022
In our second interpretability paper, we revisit “induction heads”. In 2 layer transformers these pattern-completion heads form exactly when in-context learning abruptly improves. Are they responsible for most in-context learning in large transformers? transformer-circuits.pub/202…
1
57
306
Daniela Amodei reposted
Anthropic
 @AnthropicAI
17 Feb 2022
Our first societal impacts paper explores the technical traits of large generative models and the motivations and challenges people face in building and deploying them: arxiv.org/abs/2202.07785

Predictability and Surprise in Large Generative Models

Large-scale pre-training has recently emerged as a technique for creating capable, general purpose, generative models such as GPT-3, Megatron-Turing NLG, Gopher, and many others. In this paper, we...

arxiv.org
2
33
150
Daniela Amodei reposted
Anthropic
 @AnthropicAI
22 Dec 2021
Our first interpretability paper explores a mathematical framework for trying to reverse engineer transformer language models: A Mathematical Framework for Transformer Circuits: transformer-circuits.pub/202…
3
116
612
Daniela Amodei reposted
Anthropic
 @AnthropicAI
3 Dec 2021
Our first AI alignment paper, focused on simple baselines and investigations: A General Language Assistant as a Laboratory for Alignment arxiv.org/abs/2112.00861

A General Language Assistant as a Laboratory for Alignment

Given the broad capabilities of large language models, it should be possible to work towards a general-purpose, text-based assistant that is aligned with human values, meaning that it is helpful,...

arxiv.org
5
60
322
We’re going to be focused on pushing forward our research for the next few months and are hoping to have more to share later this year. Thrilled to be working with so many talented colleagues!
7
1
28
Excited to announce what we’ve been working on this year - @AnthropicAI, an AI safety and research company. If you’d like to help us combine safety research with scaling ML models while thinking about societal impacts, check out our careers page anthropic.com/#careers
14
27
205
Daniela Amodei reposted
Anthropic
 @AnthropicAI
28 May 2021
Hello world! You can read our launch announcement here: anthropic.com/news/announcem…
5
35
297
Load more