INDEX

Explanations

Financial markets

np_max-act · gemini-2.0-flash

This neuron activates on numeric tokens—especially numbers and decimals (e.g., “250,” “0.357421875,” “0.416748046875,” etc.). It’s essentially looking for numeric data in the text.

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

 Exercise

-0.06

ріп

-0.06

_destination

-0.06

bff

-0.06

 Coverage

-0.06

 konusu

-0.06

 Shed

-0.06

 Required

-0.06

 delimiter

-0.06

POSITIVE LOGITS

ีค

0.07

Bes

0.06

Jur

0.06

に

0.06

 wieder

0.06

Env

0.06

 заліз

0.06

\Lib

0.06

German

0.06

み

0.06

Activations Density 0.034%

Financial markets

This neuron activates on numeric tokens—especially numbers and decimals (e.g., “250,” “0.357421875,” “0.416748046875,” etc.). It’s essentially looking for numeric data in the text.

No Comments

No Known Activations