INDEX

Explanations

Technical discussions

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æľīéĻĲ

-0.27

åĩı

-0.26

arat

-0.26

æŀģéĢŁ

-0.25

imat

-0.25

éĢıæĺİ

-0.25

 transitional

-0.25

ä»·

-0.25

 Transition

-0.25

bett

-0.24

POSITIVE LOGITS

per

0.37

ap

0.36

per

0.32

 closest

0.30

uce

0.30

closest

0.29

uled

0.29

pers

0.28

apy

0.28

.per

0.27

Activations Density 0.124%

Technical discussions

No Known Activations