INDEX

Explanations

country

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_12.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.12.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Beng

-0.29

ulative

-0.26

 Bailey

-0.26

irical

-0.25

å¡ĳ

-0.25

äºĭ

-0.25

åĩłä½į

-0.25

ç»ĺ

-0.25

hyp

-0.24

éĺ³åİ¿

-0.24

POSITIVE LOGITS

åĽ´ç»ķ

0.28

 surrounding

0.28

utr

0.27

å½Ģ

0.26

 Surround

0.26

bud

0.25

ç¿»èº«

0.25

cent

0.25

covered

0.25

_STAGE

0.25

Activations Density 0.010%

country

No Comments

No Known Activations