INDEX

Explanations

Computer language / code

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

allah

-0.28

åĨ²

-0.27

åĲĦåľ°

-0.27

åħ¨çĲĥ

-0.27

weep

-0.27

ä¹ĭæĺŁ

-0.26

 Worldwide

-0.26

Ø³ØªØ§ÙĨ

-0.26

per

-0.25

per

-0.25

POSITIVE LOGITS

 dissert

0.28

apr

0.26

 PROT

0.26

łģ

0.25

æĸ°å±ĢéĿ¢

0.25

 Consort

0.24

éĿ¢ç§¯çº¦

0.24

 biom

0.24

 Pens

0.23

ç¼ºå¤±

0.23

Activations Density 0.000%

Computer language / code

No Comments

No Known Activations