INDEX

Explanations

addressing the reader

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_10.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.10.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æŃ¤äºº

-0.29

ambre

-0.28

iid

-0.26

opo

-0.26

 needle

-0.26

woke

-0.26

ORE

-0.25

Portable

-0.25

å±

-0.24

 void

-0.24

POSITIVE LOGITS

ä¸įåĪĨ

0.28

åºĻ

0.25

 fundraiser

0.25

ç²¹

0.25

_distribution

0.25

 getResources

0.24

 standardized

0.24

åĪĨå¸ĥåľ¨

0.24

 bask

0.24

spark

0.23

Activations Density 0.004%

addressing the reader

No Comments

No Known Activations