INDEX

Explanations

Technical/academic papers

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

clud

-0.31

 webs

-0.27

bet

-0.26

-ab

-0.26

 taper

-0.25

oret

-0.25

immer

-0.25

Ð¿Ð°ÑĢ

-0.25

 sacrifice

-0.25

quet

-0.24

POSITIVE LOGITS

èħ¿éĥ¨

0.27

 Related

0.26

{@

0.26

 Newsletter

0.25

anium

0.25

atee

0.25

JV

0.25

izador

0.24

 Vend

0.24

related

0.23

Activations Density 0.002%

Technical/academic papers

No Comments

No Known Activations