INDEX

Explanations

stam

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_3/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 prosecuting

-0.07

突

-0.06

 audiences

-0.06

Whilst

-0.06

用

-0.06

_chart

-0.06

 acceptance

-0.06

 dreadful

-0.06

_engine

-0.06

 Would

-0.06

POSITIVE LOGITS

 Stam

0.08

 stamina

0.08

 Stamina

0.08

ternal

0.08

istem

0.07

stm

0.07

/banner

0.07

 Stamford

0.07

mej

0.07

miyor

0.06

Activations Density 0.015%

stam

No Comments

No Known Activations