INDEX

Explanations

academic texts

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

.hasNext

-0.27

fte

-0.27

èŁĴ

-0.26

 peer

-0.26

èĻļ

-0.25

 fast

-0.25

peer

-0.25

tes

-0.25

 empty

-0.24

ully

-0.24

POSITIVE LOGITS

çŃ·åŃĲ

0.27

ignon

0.26

Deb

0.26

blings

0.25

ColorBrush

0.24

 infl

0.24

åĽ½éĻħåľ¨çº¿

0.24

åħ¥èĤ¡

0.23

ownt

0.23

saida

0.23

Activations Density 0.003%

academic texts

No Known Activations