INDEX

Explanations

declines to comment

np_max-act · gemini-2.0-flash

This neuron activates on reporting‐style attribution phrases (e.g. “declined to comment,” “said,” “referring reporters,” “continued,” etc.) that introduce or qualify a speaker’s statement.

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

Interview

-0.07

,total

-0.06

 Essentially

-0.06

-ch

-0.06

ocommerce

-0.06

eral

-0.06

 Eventually

-0.06

對

-0.06

 Pill

-0.06

 її

-0.05

POSITIVE LOGITS

.Value

0.08

 passer

0.07

}->{

0.07

 Hole

0.07

 Dayton

0.06

 valve

0.06

 nová

0.06

 брат

0.06

.Provider

0.06

 posture

0.06

Activations Density 0.012%

declines to comment

This neuron activates on reporting‐style attribution phrases (e.g. “declined to comment,” “said,” “referring reporters,” “continued,” etc.) that introduce or qualify a speaker’s statement.

No Comments

No Known Activations