INDEX

Explanations

files

np_max-act · gemini-2.0-flash

The neuron triggers on French phrases that talk about processing or iterating over all files in a directory (e.g. “tous les fichiers,” “les fichiers dans…”).

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

 회원

-0.06

 takový

-0.06

 Abdul

-0.06

.gridColumn

-0.06

 unrecognized

-0.06

 openings

-0.06

_frequency

-0.06

specialchars

-0.06

 grounding

-0.06

 guitarist

-0.06

POSITIVE LOGITS

argument

0.07

 couleur

0.07

ester

0.07

 něj

0.07

 nedost

0.06

澳

0.06

 Infinity

0.06

 yüz

0.06

蒙

0.06

Aug

0.06

Activations Density 0.050%

files

The neuron triggers on French phrases that talk about processing or iterating over all files in a directory (e.g. “tous les fichiers,” “les fichiers dans…”).

No Comments

No Known Activations