INDEX

Explanations

intensifiers

np_max-act · gemini-2.0-flash

This neuron fires on degree‐or intensity-modifying words—i.e. terms used to amp up or qualify descriptions.

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

out

-0.07

split

-0.07

šetření

-0.07

trys

-0.07

�

-0.06

갖

-0.06

etc

-0.06

-out

-0.06

 plane

-0.06

.band

-0.06

POSITIVE LOGITS

 aber

0.07

整个

0.07

 znovu

0.06

 observable

0.06

 neoliberal

0.06

ImageSharp

0.06

 увер

0.06

 اختل

0.06

 Saying

0.06

fmt

0.06

Activations Density 0.271%

intensifiers

This neuron fires on degree‐or intensity-modifying words—i.e. terms used to amp up or qualify descriptions.

No Comments

No Known Activations

intensifiers

This neuron fires on degree‐or intensity-modifying words—i.e. terms used to amp up or qualify descriptions.

No Comments

No Known Activations