INDEX

Explanations

period

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 bespoke

-0.27

eming

-0.25

sty

-0.25

styl

-0.25

/umd

-0.24

 Erdogan

-0.24

ports

-0.24

 pedig

-0.24

 minimise

-0.24

 TORT

-0.24

POSITIVE LOGITS

**)

0.27

ä¸¤ç§į

0.25

 Â«

0.25

è¢Ń

0.24

We

0.24

Â«

0.24

 getWidth

0.24

áĢĸ

0.24

alo

0.24

æīĵäºĨ

0.23

Activations Density 0.023%

period

No Known Activations