INDEX

Explanations

end

np_max-act · gemini-2.0-flash

This neuron fires on words referring to parts’ ends or edges—i.e. mentions of component “ends,” “edges,” or other physical extremities.

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

Improved

-0.07

 vidéos

-0.07

August

-0.07

UN

-0.07

osate

-0.07

CUSTOM

-0.06

ivan

-0.06

 brigade

-0.06

glomer

-0.06

 landmark

-0.06

POSITIVE LOGITS

="../../

0.06

.real

0.06

nett

0.06

 Athen

0.06

 commute

0.06

(?)

0.06

以为

0.06

олн

0.06

交流

0.06

.amazon

0.06

Activations Density 0.035%

end

This neuron fires on words referring to parts’ ends or edges—i.e. mentions of component “ends,” “edges,” or other physical extremities.

No Comments

No Known Activations