INDEX

Explanations

suggestions and recommendations

np_max-act · gemini-2.0-flash

The neuron responds to normative or prescriptive language—words and phrases (like “should,” “proper,” “must,” “be”) that express how things ought to be done.

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

 proxy

-0.08

ukkit

-0.07

_zone

-0.07

 stroll

-0.07

 Adler

-0.07

Zero

-0.06

し

-0.06

GEN

-0.06

gen

-0.06

.over

-0.06

POSITIVE LOGITS

 ident

0.07

CHASE

0.06

ीं,

0.06

 تخصصی

0.06

 мають

0.06

 homeland

0.06

AU

0.06

qr

0.06

uae

0.06

 Cologne

0.06

Activations Density 0.222%

suggestions and recommendations

The neuron responds to normative or prescriptive language—words and phrases (like “should,” “proper,” “must,” “be”) that express how things ought to be done.

No Comments

No Known Activations