INDEX

Explanations

British Empire

np_max-act · gemini-2.0-flash

The neuron activates on mentions of British honourific titles and awards (e.g., appointments to the Order of the British Empire).

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

_word

-0.07

odi

-0.06

 downstream

-0.06

vh

-0.06

amy

-0.06

 obsah

-0.06

人の

-0.06

esan

-0.06

 swapping

-0.06

ocity

-0.06

POSITIVE LOGITS

__',

0.06

 그녀

0.06

 precautions

0.06

술

0.06

 Kevin

0.06

 parcels

0.06

 KeyboardInterrupt

0.06

 серьез

0.06

 espresso

0.06

_BREAK

0.06

Activations Density 0.005%

British Empire

The neuron activates on mentions of British honourific titles and awards (e.g., appointments to the Order of the British Empire).

No Comments

No Known Activations