INDEX

Explanations

code and URLs

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

usa

-0.26

 luáºŃn

-0.25

 Washer

-0.25

/wiki

-0.24

esc

-0.24

 thá»¥

-0.23

ziej

-0.23

quat

-0.23

cie

-0.23

uaÃ§Ã£o

-0.23

POSITIVE LOGITS

çĽ´åĪ°

0.28

Until

0.27

çĭĲ

0.26

çŃīåĪ°

0.26

 until

0.25

sticks

0.25

Ħä»¶

0.25

ä¿Ĺ

0.24

åĳ¼

0.24

tap

0.24

Activations Density 0.088%

code and URLs

No Comments

No Known Activations