INDEX

Explanations

technical/scientific texts

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_14.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.14.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 toolbox

-0.30

Ã¼hr

-0.28

 destination

-0.28

iations

-0.26

åľŁåľ°

-0.26

 nome

-0.25

.land

-0.25

æ³¡

-0.25

='.

-0.24

olo

-0.24

POSITIVE LOGITS

ais

0.27

anmar

0.25

LEE

0.25

ä¹Łå¥½

0.25

anted

0.24

ayas

0.24

aghan

0.24

.assertIsNot

0.23

rev

0.23

 Bing

0.23

Activations Density 0.108%

technical/scientific texts

No Known Activations