INDEX

Explanations

Computer/programming related text

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_30.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.30.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

çĻ¾å¹´

-0.29

MQ

-0.26

InstanceState

-0.26

ä¹Ŀåįģ

-0.26

 fifty

-0.26

éĿ¢

-0.26

UEST

-0.25

ered

-0.25

uraa

-0.25

utation

-0.24

POSITIVE LOGITS

ptime

0.34

çļĦæĹ¶éĹ´

0.30

æĹ¶éĹ´åĴĮ

0.29

æįŁå®³

0.28

time

0.28

gun

0.28

 Rican

0.28

 time

0.28

æĹ¶éĹ´

0.27

Time

0.27

Activations Density 0.001%

Computer/programming related text

No Known Activations