INDEX

Explanations

Alban

np_max-act · gemini-2.0-flash

This neuron responds to occurrences of the substring “Alban” (as in “Albanian” or “Albania”), i.e. references to Albanian.

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

 Need

-0.07

 Janeiro

-0.07

위를

-0.07

 resultado

-0.06

 Japan

-0.06

ège

-0.06

.percent

-0.06

Kick

-0.06

 assistir

-0.06

 Sony

-0.06

POSITIVE LOGITS

 Kosovo

0.07

WithIdentifier

0.07

 disagrees

0.07

/Subthreshold

0.06

ór

0.06

的时候

0.06

 terug

0.06

 composition

0.06

hardt

0.06

 httpResponse

0.06

Activations Density 0.008%

Alban

This neuron responds to occurrences of the substring “Alban” (as in “Albanian” or “Albania”), i.e. references to Albanian.

No Comments

No Known Activations