INDEX

Explanations

with

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

æ¢¯

-0.28

æ¶Īè´¹

-0.26

æī¹æ¬¡

-0.25

ored

-0.25

ORS

-0.24

asury

-0.24

Ð¾ÑĩÐºÐ°

-0.24

 contra

-0.24

 charged

-0.24

 consumer

-0.24

POSITIVE LOGITS

åħ¶ä»ĸäºº

0.48

ä»ĸäºº

0.41

åħ¶ä»ĸ

0.41

åħ¶ä»ĸçİ©å®¶

0.39

 others

0.39

 other

0.36

ä»»ä½ķäºº

0.36

 Others

0.35

others

0.35

æĪĸèĢħåħ¶ä»ĸ

0.35

Activations Density 0.001%

with

No Comments

No Known Activations