INDEX

Explanations

importance

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

Negative Logits

Earth

-0.29

angement

-0.28

åĲİéĿ¢

-0.27

 Controls

-0.26

å®¦

-0.26

ãĤĪãģı

-0.26

RenderingContext

-0.26

/copyleft

-0.26

Authorities

-0.25

 controls

-0.25

POSITIVE LOGITS

èº«å½±

0.34

-await

0.26

-offs

0.25

ideon

0.25

åĢ¡è®®

0.25

 selves

0.25

pants

0.25

lÃ¼

0.25

ToDevice

0.24

çļĦèº«å½±

0.24

Activations Density 0.014%

importance

No Known Activations