INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron activates on CSS style declarations (the “property: value” tokens such as property names, the colon, and the value).

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

 등을

-0.06

topics

-0.06

<t

-0.06

vez

-0.06

 trem

-0.06

PointerException

-0.06

관계

-0.06

�

-0.06

 SOAP

-0.06

Expired

-0.06

POSITIVE LOGITS

 flavours

0.07

_board

0.06

-heavy

0.06

 voll

0.06

 Northeast

0.06

 flavors

0.06

イ

0.06

Ф

0.06

'];?>↵

0.06

Activations Density 0.073%

code

The neuron activates on CSS style declarations (the “property: value” tokens such as property names, the colon, and the value).

No Comments

No Known Activations