INDEX

Explanations

science and technology

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

ä»»

-0.27

San

-0.26

å¼ºè¿«

-0.25

ëŀĮ

-0.25

çķ

-0.25

åĪ¥äºº

-0.24

æľªç»ı

-0.24

OutputStream

-0.24

èĩªçĦ¶çģ¾å®³

-0.24

Ø¬ÙĬÙĨ

-0.24

POSITIVE LOGITS

æľįåĬ¡è´¨éĩı

0.27

undi

0.26

 reckon

0.26

 ná»¯a

0.25

 today

0.25

AGER

0.25

æł¸å®ļ

0.25

 nests

0.25

hic

0.25

loser

0.24

Activations Density 0.041%

science and technology

No Comments

No Known Activations