INDEX

Explanations

onto

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CTYPE

-0.28

çķĮ

-0.28

ä¸ĬéĹ¨

-0.27

 clap

-0.27

è¿Ł

-0.26

éĨ

-0.25

ä¸Ģåĳ³

-0.25

chemy

-0.25

åĶĽ

-0.24

æİĴ

-0.24

POSITIVE LOGITS

ROLLER

0.30

-spot

0.29

adder

0.29

 fixture

0.26

 acess

0.26

allery

0.26

åħ¨æł¡

0.25

-site

0.25

ellation

0.25

isure

0.24

Activations Density 0.164%

onto

No Comments

No Known Activations