INDEX

Explanations

win

np_max-act · gemini-2.0-flash

The main thing this neuron does is spot contest and giveaway language (e.g. mentions of “win,” “enter,” “raffle,” or prize‐related calls to action).

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

 propaganda

-0.06

Lines

-0.06

_sequences

-0.06

-stream

-0.06

 experiments

-0.06

Load

-0.06

(D

-0.06

Default

-0.06

 lovers

-0.06

fad

-0.06

POSITIVE LOGITS

 потрап

0.07

 Locker

0.07

 τρό

0.07

odium

0.07

ционной

0.07

 країн

0.06

 deton

0.06

 контр

0.06

_hdl

0.06

 ملي

0.06

Activations Density 0.012%

win

The main thing this neuron does is spot contest and giveaway language (e.g. mentions of “win,” “enter,” “raffle,” or prize‐related calls to action).

No Comments

No Known Activations