INDEX

Explanations

keen

np_max-act · gemini-2.0-flash

This neuron fires on language describing sharp or heightened perception (e.g. “keen sense,” “piercing eyes”).

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

amburger

-0.07

.functional

-0.07

 anders

-0.06

apas

-0.06

 전체

-0.06

 functionName

-0.06

oras

-0.06

}}}

-0.06

实现

-0.06

anging

-0.06

POSITIVE LOGITS

 informed

0.07

 heightened

0.07

 keen

0.07

----↵

0.06

SUN

0.06

İN

0.06

 agreed

0.06

 hone

0.06

 dean

0.06

witter

0.06

Activations Density 0.006%

keen

This neuron fires on language describing sharp or heightened perception (e.g. “keen sense,” “piercing eyes”).

No Comments

No Known Activations