INDEX

Explanations

um

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

undy

-0.31

 cylinders

-0.29

loys

-0.28

rib

-0.28

å¼¥

-0.27

 scarf

-0.27

æīĺ

-0.26

çŃĴ

-0.26

 Doug

-0.26

éĦ±

-0.26

POSITIVE LOGITS

ë¹Ĺ

0.34

ä»¥èī²

0.28

 Ð´ÐµÐ»Ðµ

0.27

 ÑģÐ²ÑıÐ·Ð°Ð½

0.26

 BITS

0.26

eel

0.25

æ²«

0.25

çħİ

0.24

ONSE

0.24

äºĴåĬ¨

0.24

Activations Density 0.003%

um

No Comments

No Known Activations