INDEX

Explanations

Se

np_max-act · gemini-2.0-flash

The neuron activates on Cyrillic‐script text (i.e. Russian words or letters).

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

 Fraser

-0.06

 save

-0.06

 forg

-0.06

/P

-0.06

-Bold

-0.06

�

-0.06

#================================================================

-0.06

_exec

-0.06

 Marg

-0.06

>().

-0.06

POSITIVE LOGITS

 раздел

0.07

 respective

0.07

 circumcision

0.07

INCLUDED

0.07

 Stephanie

0.06

_PROTOCOL

0.06

 także

0.06

ава

0.06

 Economist

0.06

 cool

0.06

Activations Density 0.020%

Se

The neuron activates on Cyrillic‐script text (i.e. Russian words or letters).

No Comments

No Known Activations