INDEX

Explanations

pi

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

dig

-0.29

 largo

-0.27

cept

-0.27

east

-0.26

bt

-0.26

æĸ°é²ľ

-0.25

åĬ¨æĢģ

-0.25

ä»Ģä¹Īäºĭæĥħ

-0.25

ghi

-0.24

dynamic

-0.24

POSITIVE LOGITS

 kick

0.34

æį»

0.26

 bleed

0.26

.Emit

0.25

 kicked

0.24

è¯·ä½ł

0.24

kick

0.24

æĸĻ

0.24

undler

0.24

ãģ¿

0.23

Activations Density 0.003%

pi

No Comments

No Known Activations