INDEX

Explanations

possessive

np_max-act · gemini-2.0-flash

The neuron primarily detects the possessive suffix "'s" attached to words.

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

Endpoints

-0.06

appendTo

-0.06

 lực

-0.06

 defaultMessage

-0.06

 Fighters

-0.06

ウォ

-0.06

amaha

-0.06

…↵↵↵↵

-0.06

 кле

-0.06

POSITIVE LOGITS

 urgent

0.08

Right

0.07

(rank

0.07

 veriyor

0.06

 наш

0.06

stdexcept

0.06

 lavish

0.06

[ix

0.06

"&

0.06

Activations Density 0.022%

possessive

The neuron primarily detects the possessive suffix "'s" attached to words.

No Comments

No Known Activations