INDEX

Explanations

="

np_max-act · gemini-2.0-flash

The neuron activates on XML namespace‐URL tokens (e.g. parts of “xmlns=„http://…/POM/4.0.0””) in Maven POM files.

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

:@

-0.07

_STRUCTURE

-0.07

-0.06

["@

-0.06

мерик

-0.06

decimal

-0.06

.album

-0.06

otion

-0.06

�

-0.06

_SQL

-0.06

POSITIVE LOGITS

0.06

_lead

0.06

 seems

0.06

 मई

0.06

 mounted

0.06

ór

0.06

 клас

0.06

inator

0.06

 signs

0.06

 Shuttle

0.06

Activations Density 0.001%

="

The neuron activates on XML namespace‐URL tokens (e.g. parts of “xmlns=„http://…/POM/4.0.0””) in Maven POM files.

No Comments

No Known Activations