INDEX

Explanations

Actions and results

np_max-act · gemini-2.0-flash

This neuron activates on past-participle verb forms (e.g., “achieved,” “delivered,” “provided,” “accomplished”).

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

-paid

-0.07

god

-0.07

 realizado

-0.07

	packet

-0.06

.origin

-0.06

 yaparak

-0.06

 proportion

-0.06

 müşteri

-0.06

anceled

-0.06

 wich

-0.06

POSITIVE LOGITS

Ý

0.07

.multi

0.07

呀

0.06

\C

0.06

 funkc

0.06

 piercing

0.06

■

0.06

 نمی

0.06

 assms

0.06

 miserable

0.06

Activations Density 0.239%

Actions and results

This neuron activates on past-participle verb forms (e.g., “achieved,” “delivered,” “provided,” “accomplished”).

No Comments

No Known Activations

Actions and results

This neuron activates on past-participle verb forms (e.g., “achieved,” “delivered,” “provided,” “accomplished”).

No Comments

No Known Activations