INDEX

Explanations

place names

np_max-act · gemini-2.0-flash

The neuron responds to letters with diacritics (accented or non-ASCII characters) in place names.

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

departments

-0.06

 прод

-0.06

 Wahl

-0.06

pection

-0.06

.dtd

-0.06

templates

-0.06

Synopsis

-0.06

>Main

-0.06

 federation

-0.06

.getLogin

-0.05

POSITIVE LOGITS

 lady

0.07

_eps

0.07

folder

0.07

 girl

0.06

_os

0.06

 libr

0.06

cae

0.06

/>\

0.06

 organic

0.06

Activations Density 0.022%

place names

The neuron responds to letters with diacritics (accented or non-ASCII characters) in place names.

No Comments

No Known Activations

place names

The neuron responds to letters with diacritics (accented or non-ASCII characters) in place names.

No Comments

No Known Activations