INDEX

Explanations

citation numbers

np_max-act · gemini-2.0-flash

This neuron is triggered by numeric tokens, i.e. sequences of digits (e.g. years, grant or docket numbers, citation IDs).

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

นอ

-0.07

contributors

-0.07

 вик

-0.07

-prop

-0.06

removeAttr

-0.06

Open

-0.06

 blush

-0.06

.splitext

-0.06

Alt

-0.06

-Version

-0.06

POSITIVE LOGITS

{%

0.07

 stránky

0.06

 hassle

0.06

ỗi

0.06

(tile

0.06

 veterinary

0.06

bei

0.06

ical

0.06

Activations Density 0.041%

citation numbers

This neuron is triggered by numeric tokens, i.e. sequences of digits (e.g. years, grant or docket numbers, citation IDs).

No Comments

No Known Activations