INDEX

Explanations

rather

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æ»ļæ»ļ

-0.30

iaux

-0.29

æ»¾

-0.28

æ»ļ

-0.27

brtc

-0.26

(()

-0.25

(JS

-0.25

;\↵

-0.24

chie

-0.24

pat

-0.24

POSITIVE LOGITS

like

0.31

æ¯Ķå¦Ĥ

0.30

åĥı

0.28

å°±åĥı

0.28

 Bord

0.28

et

0.27

ä»¥äººä¸º

0.27

antages

0.26

 Like

0.26

_than

0.26

Activations Density 0.007%

rather

No Comments

No Known Activations