INDEX

Explanations

int

np_max-act · gemini-2.0-flash

This neuron consistently activates on the return‐type token in Java method declarations.

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

HID

-0.07

_PR

-0.07

TON

-0.06

label

-0.06

 Search

-0.06

 UICollectionView

-0.06

_FT

-0.06

"."

-0.06

 preceded

-0.06

 sequel

-0.06

POSITIVE LOGITS

 작은

0.06

}{

0.06

になった

0.06

 último

0.06

ampilkan

0.06

 Εκ

0.06

 الل

0.06

аем

0.06

 niż

0.06

 nackte

0.06

Activations Density 0.007%

int

This neuron consistently activates on the return‐type token in Java method declarations.

No Comments

No Known Activations