INDEX

Explanations

asterisk

np_max-act · gemini-2.0-flash

The neuron consistently fires on the asterisk symbol (“*”), spotting formatting or footnote markers in technical or tabular text.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 batt

-0.07

_MIN

-0.07

rus

-0.07

DJ

-0.07

 Hond

-0.07

등

-0.06

İY

-0.06

(AP

-0.06

 contin

-0.06

_min

-0.06

POSITIVE LOGITS

+'/

0.06

allel

0.06

lico

0.06

+"_

0.06

\Response

0.06

anoi

0.06

failure

0.06

ис

0.06

emma

0.06

ichick

0.05

Activations Density 0.001%

asterisk

The neuron consistently fires on the asterisk symbol (“*”), spotting formatting or footnote markers in technical or tabular text.

No Comments

No Known Activations