INDEX

Explanations

sibling

np_max-act · gemini-2.0-flash

This neuron activates on the word “sibling” (and its variants, including plural or stem forms).

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nearer

-0.07

frage

-0.06

 letras

-0.06

%.

-0.06

 Porter

-0.06

_trajectory

-0.06

 prostřed

-0.06

_Des

-0.06

 Quang

-0.06

Way

-0.06

POSITIVE LOGITS

 sibling

0.13

 siblings

0.13

 Family

0.08

 tüm

0.07

 сви

0.07

ibling

0.07

 child

0.07

 brawl

0.07

 family

0.07

 pervasive

0.06

Activations Density 0.002%

sibling

This neuron activates on the word “sibling” (and its variants, including plural or stem forms).

No Comments

No Known Activations