INDEX

Explanations

batteries

np_max-act · gemini-2.0-flash

This neuron specifically detects the Polish word “baterii” (the inflected form of “bateria,” meaning “battery”).

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

タル

-0.06

 destroys

-0.06

 piè

-0.06

 eighth

-0.06

 suicide

-0.06

 barrel

-0.06

.Collection

-0.06

 blas

-0.06

ocide

-0.06

-upload

-0.06

POSITIVE LOGITS

％

0.07

ukarı

0.07

 Canada

0.07

เค

0.07

ライト

0.07

 responder

0.06

 sake

0.06

 resolving

0.06

 الجز

0.06

 desar

0.06

Activations Density 0.008%

batteries

This neuron specifically detects the Polish word “baterii” (the inflected form of “bateria,” meaning “battery”).

No Comments

No Known Activations

batteries

This neuron specifically detects the Polish word “baterii” (the inflected form of “bateria,” meaning “battery”).

No Comments

No Known Activations