INDEX

Explanations

Tech companies

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

èľķ

-0.31

 site

-0.26

 crem

-0.26

èĿ¶

-0.26

 orgas

-0.25

Stra

-0.24

è¡Ķæİ¥

-0.24

æł¡å¤ĸ

-0.24

_eg

-0.24

site

-0.24

POSITIVE LOGITS

åĳĺå·¥

0.34

 Fiber

0.33

 stream

0.30

_stream

0.29

 streamed

0.28

Streamer

0.28

 engineers

0.27

å·¥ç¨ĭå¸Ī

0.27

 Stream

0.27

stream

0.27

Activations Density 0.943%

Tech companies

No Known Activations