INDEX

Explanations

punctuation

np_max-act · gemini-2.0-flash

This neuron detects subjective, evaluative words—especially opinionated adjectives and nouns (e.g. “essential,” “absurd,” “charlatans,” “naive”) that convey a writer’s stance or judgment.

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

-0.06

 Bakan

-0.06

ロン

-0.06

_ALIAS

-0.06

NH

-0.06

 законодавства

-0.06

 deduction

-0.06

 Nicholson

-0.06

ertoire

-0.05

_apps

-0.05

POSITIVE LOGITS

(MAX

0.07

usb

0.07

fir

0.07

=:

0.06

ไ

0.06

 Sidebar

0.06

,state

0.06

 brown

0.06

 Farmers

0.06

_TAGS

0.06

Activations Density 0.056%

punctuation

This neuron detects subjective, evaluative words—especially opinionated adjectives and nouns (e.g. “essential,” “absurd,” “charlatans,” “naive”) that convey a writer’s stance or judgment.

No Comments

No Known Activations