INDEX

Explanations

special characters

np_max-act · gemini-2.0-flash

The neuron selectively activates on numeric literals (especially floating‐point numbers) in code.

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

 brightness

-0.07

事業

-0.07

 Draw

-0.07

 classrooms

-0.07

 Kart

-0.07

ustom

-0.06

 border

-0.06

.completed

-0.06

manı

-0.06

LE

-0.06

POSITIVE LOGITS

それは

0.07

บท

0.06

 Riot

0.06

mpi

0.06

-logo

0.06

 게시물

0.06

 Москва

0.06

 зависим

0.06

Activations Density 0.021%

special characters

The neuron selectively activates on numeric literals (especially floating‐point numbers) in code.

No Comments

No Known Activations

special characters

The neuron selectively activates on numeric literals (especially floating‐point numbers) in code.

No Comments

No Known Activations