INDEX

Explanations

membrane

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åħļåĴĮåĽ½å®¶

-0.30

Wis

-0.28

Har

-0.27

 ((((

-0.26

(({

-0.26

 Lucia

-0.25

ä¸Ģè§Ĵ

-0.25

DNC

-0.25

 Herc

-0.24

ç²¾å¿ĥ

-0.24

POSITIVE LOGITS

 being

0.42

being

0.35

 having

0.29

 Ã©tant

0.28

itas

0.28

 Qualified

0.27

åĬ¡

0.27

urch

0.26

having

0.26

 qualified

0.25

Activations Density 0.002%

membrane

No Comments

No Known Activations