INDEX

Explanations

ch

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_35.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.35.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æİı

-0.27

 piping

-0.26

escape

-0.26

 puss

-0.26

EDITOR

-0.26

æİĴè¡Į

-0.26

 escape

-0.25

åºĸ

-0.25

Bun

-0.25

 altogether

-0.25

POSITIVE LOGITS

æĺİæĺ¾

0.30

onds

0.29

ungeon

0.29

ond

0.28

æĥ³

0.28

à¹Ģà¸Ķ

0.27

ç®Ģåįķ

0.27

ventory

0.26

monds

0.26

ëł¨

0.25

Activations Density 0.058%

ch

No Known Activations