INDEX

Explanations

inclination

np_max-act · gemini-2.0-flash

The neuron detects words describing an object’s tilt or inclination (e.g. “tilted,” “equator,” “axis,” “inclination”).

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

cores

-0.07

يز

-0.07

 Unix

-0.06

ยน

-0.06

 \\↵

-0.06

Baseline

-0.06

 businesses

-0.06

 normalization

-0.06

 organisations

-0.06

وق

-0.06

POSITIVE LOGITS

ast

0.07

_FAR

0.07

.RELATED

0.06

@Data

0.06

Forgot

0.06

 söylem

0.06

sak

0.06

sui

0.06

.addItem

0.06

(HWND

0.06

Activations Density 0.002%

inclination

The neuron detects words describing an object’s tilt or inclination (e.g. “tilted,” “equator,” “axis,” “inclination”).

No Comments

No Known Activations