INDEX

Explanations

transcript/dialogue

np_max-act · gemini-2.0-flash

This neuron activates on the “Agent:” speaker-turn indicator, i.e. it detects when the agent is about to speak.

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

 night

-0.07

 огля

-0.07

NA

-0.07

(Schedulers

-0.07

 fifty

-0.07

 nanoparticles

-0.06

Yahoo

-0.06

 chor

-0.06

 enthusiasts

-0.06

 Xiaomi

-0.06

POSITIVE LOGITS

 Đặc

0.07

malink

0.06

 bastante

0.06

гар

0.06

ιχ

0.06

αρά

0.06

も

0.06

거

0.06

탄

0.06

esel

0.06

Activations Density 0.019%

transcript/dialogue

This neuron activates on the “Agent:” speaker-turn indicator, i.e. it detects when the agent is about to speak.

No Comments

No Known Activations