INDEX

Explanations

assault/attack

np_max-act · gemini-2.0-flash

This neuron detects verbs and terms describing violent or defiant actions—especially assaulting, resisting, obstructing, or fleeing from law enforcement.

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

 lettre

-0.07

グラ

-0.07

dbcTemplate

-0.07

액

-0.06

 militar

-0.06

[channel

-0.06

 uterus

-0.06

 средства

-0.06

dbg

-0.06

PL

-0.06

POSITIVE LOGITS

 Newsletter

0.07

 Adopt

0.07

��

0.07

 Derby

0.06

 adopt

0.06

dep

0.06

 deprivation

0.06

_$

0.06

 distr

0.06

anc

0.06

Activations Density 0.012%

assault/attack

This neuron detects verbs and terms describing violent or defiant actions—especially assaulting, resisting, obstructing, or fleeing from law enforcement.

No Comments

No Known Activations