INDEX

Explanations

giving a bad name

np_max-act · gemini-2.0-flash

The neuron specifically detects the idiomatic expression of tarnishing a reputation—most strongly “give(s) X a bad name” or variants like “spoil Y for everyone.”

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

กำหนด

-0.07

 золот

-0.06

valu

-0.06

答

-0.06

的事

-0.06

pth

-0.06

 SIGN

-0.06

 spiritual

-0.06

ート

-0.06

 Talent

-0.05

POSITIVE LOGITS

 insp

0.07

_undo

0.07

 정규

0.07

(infile

0.07

auled

0.07

.vx

0.07

ेस

0.06

↔

0.06

.cond

0.06

.Reverse

0.06

Activations Density 0.039%

giving a bad name

The neuron specifically detects the idiomatic expression of tarnishing a reputation—most strongly “give(s) X a bad name” or variants like “spoil Y for everyone.”

No Comments

No Known Activations