INDEX

Explanations

publish/published

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_24.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.24.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ç¤¾ä¼ļèµĦæľ¬

-0.27

ãģ©ãģĨãģĹãģ¦ãĤĤ

-0.26

FI

-0.24

ivid

-0.24

 arms

-0.24

 classifications

-0.24

ç¦»

-0.24

 references

-0.23

unused

-0.23

äºĮåįģå¹´

-0.23

POSITIVE LOGITS

ubb

0.30

æĪĲäºº

0.27

aÅĤ

0.25

è¿ĽæŃ¥

0.24

-placeholder

0.24

çļĦæĦŁæĥħ

0.24

bots

0.24

ourn

0.23

emergency

0.23

anes

0.23

Activations Density 0.056%

publish/published

No Known Activations