INDEX

Explanations

scientific publications and patents

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

references to academic or technical document identifiers, particularly in arxiv-style preprint formats.

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

This neuron detects academic archive category tags and identifiers (e.g. “hep-ph”, “hep-th”, “cond-mat”, “comp-lang”, etc.).

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

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

 that

-1.23

withCredentials

-1.16

Ủ

-1.09

.”

-1.04

 schlaf

-1.04

ôtel

-1.02

 advir

-1.01

 exigir

-1.00

arctic

-0.99

áne

-0.98

POSITIVE LOGITS

茈

1.31

 تھے

1.23

Elő

1.19

ться

1.13

 گیا

1.11

 ہیں۔

1.04

 entscheid

1.04

ございました

1.03

 あん

1.02

９

1.02

Activations Density 0.002%

scientific publications and patents

references to academic or technical document identifiers, particularly in arxiv-style preprint formats.

This neuron detects academic archive category tags and identifiers (e.g. “hep-ph”, “hep-th”, “cond-mat”, “comp-lang”, etc.).

No Comments

No Known Activations

scientific publications and patents

references to academic or technical document identifiers, particularly in arxiv-style preprint formats.

This neuron detects academic archive category tags and identifiers (e.g. “hep-ph”, “hep-th”, “cond-mat”, “comp-lang”, etc.).

No Comments

No Known Activations