INDEX

Explanations

ular

np_max-act · gemini-2.0-flash

The neuron is triggered by the character sequence “ular,” i.e. words ending in the suffix “ular.”

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

 hardly

-0.07

-0.06

Rol

-0.06

 đáo

-0.06

 Parm

-0.06

 دان

-0.06

 gums

-0.06

 unjust

-0.06

Mb

-0.06

Boy

-0.06

POSITIVE LOGITS

ék

0.08

团

0.07

utor

0.06

 команд

0.06

ác

0.06

serie

0.06

 вико

0.06

ститут

0.06

砲

0.06

اريخ

0.06

Activations Density 0.001%

ular

The neuron is triggered by the character sequence “ular,” i.e. words ending in the suffix “ular.”

No Comments

No Known Activations