INDEX

Explanations

conversations

np_max-act · gemini-2.0-flash

This neuron responds to common function words (short, high-frequency tokens like “a,” “and,” “while,” “Sit,” “dirty,” “hands,” “clean”) that often appear in connected phrases or imperatives.

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

то

-0.06

 Bing

-0.06

 Investigators

-0.06

 Oswald

-0.06

Blockly

-0.06

FTP

-0.06

 eighty

-0.06

ODB

-0.06

 الانت

-0.06

POSITIVE LOGITS

 estoy

0.06

 Replica

0.06

AsyncResult

0.06

ippines

0.06

DMIN

0.06

hle

0.06

 koşul

0.06

 replica

0.06

.Matcher

0.06

교회

0.06

Activations Density 0.125%

conversations

This neuron responds to common function words (short, high-frequency tokens like “a,” “and,” “while,” “Sit,” “dirty,” “hands,” “clean”) that often appear in connected phrases or imperatives.

No Comments

No Known Activations

conversations

This neuron responds to common function words (short, high-frequency tokens like “a,” “and,” “while,” “Sit,” “dirty,” “hands,” “clean”) that often appear in connected phrases or imperatives.

No Comments

No Known Activations