Zobrazeno 1 - 10
of 287
pro vyhledávání: '"Smith, Logan A."'
Measuring Progress in Dictionary Learning for Language Model Interpretability with Board Game Models
Autor:
Karvonen, Adam, Wright, Benjamin, Rager, Can, Angell, Rico, Brinkmann, Jannik, Smith, Logan, Verdun, Claudio Mayrink, Bau, David, Marks, Samuel
What latent features are encoded in language model (LM) representations? Recent work on training sparse autoencoders (SAEs) to disentangle interpretable features in LM representations has shown significant promise. However, evaluating the quality of
Externí odkaz:
http://arxiv.org/abs/2408.00113
Autor:
Belrose, Nora, Furman, Zach, Smith, Logan, Halawi, Danny, Ostrovsky, Igor, McKinney, Lev, Biderman, Stella, Steinhardt, Jacob
We analyze transformers from the perspective of iterative inference, seeking to understand how model predictions are refined layer by layer. To do so, we train an affine probe for each block in a frozen pretrained model, making it possible to decode
Externí odkaz:
http://arxiv.org/abs/2303.08112
AI alignment research is the field of study dedicated to ensuring that artificial intelligence (AI) benefits humans. As machine intelligence gets more advanced, this research is becoming increasingly important. Researchers in the field share ideas ac
Externí odkaz:
http://arxiv.org/abs/2206.02841
Publikováno v:
In Journal of Affective Disorders 1 September 2024 360:33-41
Publikováno v:
In Neuroscience and Biobehavioral Reviews March 2024 158
Autor:
Smith, Logan A., Hicks, Illya V.
To monitor electrical activity throughout the power grid and mitigate outages, sensors known as phasor measurement units can installed. Due to implementation costs, it is desirable to minimize the number of sensors deployed while ensuring that the gr
Externí odkaz:
http://arxiv.org/abs/2006.03460
We present an integer programming model to compute the strong rainbow connection number, $src(G)$, of any simple graph $G$. We introduce several enhancements to the proposed model, including a fast heuristic, and a variable elimination scheme. Moreov
Externí odkaz:
http://arxiv.org/abs/2006.02988
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.