INDEX

Explanations

numerical values and units

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

 vent

-0.31

åıĳ

-0.31

ivel

-0.30

 Vent

-0.29

éĹ·

-0.28

çĸĹæ³ķ

-0.27

letic

-0.27

éĩĬæĶ¾

-0.27

 Mobil

-0.26

 release

-0.25

POSITIVE LOGITS

cid

0.27

 ignoring

0.26

 entertained

0.25

look

0.25

 ÑĥÐºÑĢÐ°

0.24

vis

0.24

èĻ¹

0.24

 Late

0.24

 demon

0.23

OrNull

0.23

Activations Density 0.002%

numerical values and units

No Known Activations