INDEX

Explanations

The neuron selectively activates on numeric literals (sequences of digits) in the text.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

namespace syntax, names, and scientific/technical terms

np_acts-logits-general · gemini-2.5-flash-lite

mobile app and software development related terminology.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 아니라

-1.26

 після

-1.13

 glorious

-1.10

んじゃない

-1.09

Werdegang

-1.08

 очень

-1.05

 дуже

-1.05

()];

-1.02

 enormous

-1.02

 rất

-1.02

POSITIVE LOGITS

ног

1.09

 oreja

1.05

 favoriser

1.05

 zwią

1.04

 широкий

1.02

údos

1.01

COB

0.99

 domaines

0.99

IDOS

0.98

 Diseño

0.97

Activations Density 0.001%

The neuron selectively activates on numeric literals (sequences of digits) in the text.

namespace syntax, names, and scientific/technical terms

mobile app and software development related terminology.

No Comments

No Known Activations

The neuron selectively activates on numeric literals (sequences of digits) in the text.

namespace syntax, names, and scientific/technical terms

mobile app and software development related terminology.

No Comments

No Known Activations