INDEX

Explanations

Come

np_max-act · gemini-2.0-flash

This neuron detects the imperative call-to-action “Come” (and its variants) used to start invitations or announcements.

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

 inhib

-0.07

(signal

-0.07

otation

-0.06

IELDS

-0.06

 spac

-0.06

(labels

-0.06

 camps

-0.06

 프리

-0.06

�

-0.06

 binary

-0.06

POSITIVE LOGITS

 Come

0.10

Come

0.09

 come

0.08

pole

0.07

..↵↵↵↵

0.07

 diye

0.07

 serta

0.07

tout

0.07

مم

0.07

ebe

0.07

Activations Density 0.008%

Come

This neuron detects the imperative call-to-action “Come” (and its variants) used to start invitations or announcements.

No Comments

No Known Activations