INDEX

Explanations

be

np_max-act · gemini-2.0-flash

The neuron fires on non-English text fragments (e.g. subword tokens from Arabic, Italian, Spanish, etc.).

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

any

-0.07

ायन

-0.07

 useful

-0.07

 tiện

-0.06

iren

-0.06

ANY

-0.06

.`|`↵

-0.06

ydı

-0.06

ЛИ

-0.06

џ

-0.06

POSITIVE LOGITS

=?

0.07

//[

0.07

#-

0.07

_ret

0.06

ชาต

0.06

...')↵

0.06

{$

0.06

uess

0.06

SU

0.06

.complete

0.06

Activations Density 0.203%

be

The neuron fires on non-English text fragments (e.g. subword tokens from Arabic, Italian, Spanish, etc.).

No Comments

No Known Activations