INDEX

Explanations

software settings

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æ¸ħæ°´

-0.26

ienie

-0.26

Smarty

-0.25

 asteroids

-0.24

Å¼a

-0.24

×ĳ×¨

-0.24

ieÅĦ

-0.24

resents

-0.23

vice

-0.23

ierte

-0.23

POSITIVE LOGITS

ä¹ĭä½Ļ

0.31

Pet

0.28

Plus

0.26

åĪĨåī²

0.26

 Cycle

0.26

Pet

0.26

Cycle

0.25

eren

0.25

åįĩåįİ

0.25

 Plus

0.25

Activations Density 0.016%

software settings

No Comments

No Known Activations