INDEX

Explanations

)

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 aside

-0.35

inski

-0.29

 suggestion

-0.26

 suggestions

-0.26

 clue

-0.26

 overd

-0.25

tip

-0.24

si

-0.24

aside

-0.24

sky

-0.24

POSITIVE LOGITS

 operators

0.26

 Operators

0.25

[section

0.25

 mÃŃ

0.25

 organizational

0.25

[assembly

0.25

drops

0.24

 mediums

0.24

éĢł

0.24

getQuery

0.24

Activations Density 0.010%

)

No Comments

No Known Activations