INDEX

Explanations

Finding explanations / information

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

Negative Logits

ä¸ĢåľĪ

-0.25

Raise

-0.25

 Raise

-0.25

 raise

-0.25

Hands

-0.24

 Minds

-0.24

ç«Ļèµ·æĿ¥

-0.24

æ¾ľ

-0.24

 CancellationToken

-0.23

AuthToken

-0.23

POSITIVE LOGITS

 descri

0.27

<![

0.26

ä¸ĢçĤ¹

0.24

ç§ĳçĽ®

0.24

vers

0.24

 pieces

0.24

Piece

0.24

 descr

0.24

 Pieces

0.24

atically

0.23

Activations Density 0.006%

Finding explanations / information

No Known Activations