INDEX

Explanations

vampires

np_max-act · gemini-2.0-flash

The neuron is looking for “Dracula”/vampire–style roleplay cues and the surrounding header/control tokens in the prompt.

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

illary

-0.07

 ولي

-0.06

晴

-0.06

uff

-0.06

Το

-0.06

 Exec

-0.06

logged

-0.06

ались

-0.06

.life

-0.06

 اث

-0.06

POSITIVE LOGITS

0.07

 encounters

0.07

div

0.06

Ty

0.06

 dispose

0.06

.emf

0.06

NavLink

0.06

 acad

0.06

 disposal

0.06

Activations Density 0.022%

vampires

The neuron is looking for “Dracula”/vampire–style roleplay cues and the surrounding header/control tokens in the prompt.

No Comments

No Known Activations

vampires

The neuron is looking for “Dracula”/vampire–style roleplay cues and the surrounding header/control tokens in the prompt.

No Comments

No Known Activations