INDEX

Explanations

I

np_max-act · gemini-2.0-flash

This neuron activates on tokens that look like interface names (type identifiers beginning with “I” such as IRequest, IModule, IObservable, etc.).

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

 strange

-0.06

 nước

-0.06

 lends

-0.06

历史

-0.06

補

-0.06

奏

-0.06

 synthes

-0.06

 bais

-0.06

 quienes

-0.06

leton

-0.06

POSITIVE LOGITS

_Process

0.07

uffles

0.06

 Agility

0.06

 worthless

0.06

getAll

0.06

ri

0.06

losures

0.06

 Broken

0.06

imax

0.06

Activations Density 0.005%

I

This neuron activates on tokens that look like interface names (type identifiers beginning with “I” such as IRequest, IModule, IObservable, etc.).

No Comments

No Known Activations