INDEX

Explanations

Scientific/technical texts

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

legg

-0.32

harma

-0.29

xFFFF

-0.26

enschaft

-0.26

_manifest

-0.26

irms

-0.25

eman

-0.25

steen

-0.25

 cabe

-0.24

aspers

-0.24

POSITIVE LOGITS

sit

0.27

'[

0.26

istrict

0.25

æ¶ķ

0.25

bomb

0.24

æĻ¤

0.24

 Carbon

0.24

onomy

0.24

èĩ¾

0.24

ç¢³

0.24

Activations Density 24.805%

Scientific/technical texts

No Known Activations