INDEX

Explanations

example

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Throws

-0.07

(M

-0.07

大道

-0.06

铁路

-0.06

痘痘

-0.06

Filed

-0.06

/web

-0.06

etrofit

-0.06

 freopen

-0.06

社保

-0.06

POSITIVE LOGITS

/effects

0.08

 Orthodox

0.07

 według

0.07

_persona

0.07

 useContext

0.07

Fetching

0.06

Ὑ

0.06

 Dysfunction

0.06

 upholstery

0.06

 starvation

0.06

Activations Density 0.017%

example

No Comments

No Known Activations