INDEX

Explanations

best

np_max-act · gemini-2.0-flash

The neuron flags occurrences of the superlative “best” (i.e. positive highest-degree praise).

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

Channels

-0.07

margin

-0.06

 avoided

-0.06

Orders

-0.06

flux

-0.06

 home

-0.06

dyž

-0.06

 remove

-0.06

 Hawkins

-0.06

 Бог

-0.06

POSITIVE LOGITS

dsn

0.07

설

0.07

(propertyName

0.07

/\.

0.07

＼＼

0.07

UIScreen

0.07

-mult

0.06

กรรม

0.06

 hton

0.06

 ống

0.06

Activations Density 0.050%

best

The neuron flags occurrences of the superlative “best” (i.e. positive highest-degree praise).

No Comments

No Known Activations

best

The neuron flags occurrences of the superlative “best” (i.e. positive highest-degree praise).

No Comments

No Known Activations