INDEX

Explanations

place names and locations

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

ç®¡çĲĨä¸Ńå¿ĥ

-0.27

 onView

-0.27

æºĲæºĲä¸įæĸŃ

-0.27

 conceivable

-0.26

aroo

-0.26

suit

-0.25

 Nath

-0.24

Ø§Ø¦Ø²

-0.24

untlet

-0.24

nap

-0.24

POSITIVE LOGITS

åħ³

0.30

omic

0.29

Sweden

0.27

xic

0.26

åĵģ

0.26

ood

0.26

ober

0.25

åľ°ä½į

0.25

oming

0.25

Ð¾Ð´Ðµ

0.24

Activations Density 0.883%

place names and locations

No Known Activations