INDEX

Explanations

First MAX_ACTIVATING_TOKENS when other methods failith

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åĪĨå±Ģ

-0.27

éļĺ

-0.26

è¡Ķ

-0.26

press

-0.25

ä»İå°ı

-0.24

pressed

-0.24

èĴľ

-0.24

æ°´åĪĨ

-0.24

 press

-0.24

ponsored

-0.24

POSITIVE LOGITS

ç»Ŀ

0.31

çµķ

0.26

amps

0.25

asket

0.25

åĨ³

0.24

dvd

0.24

-abs

0.23

è¡¥é½Ĳ

0.23

ABS

0.23

ç»Ŀå¯¹

0.23

Activations Density 0.002%

First MAX_ACTIVATING_TOKENS when other methods failith

No Comments

No Known Activations