INDEX

Explanations

Reg

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 trace

-0.36

 drive

-0.35

 feed

-0.33

 form

-0.32

è·¡

-0.31

 plan

-0.30

 follow

-0.30

 pose

-0.30

 look

-0.30

 search

-0.30

POSITIVE LOGITS

åĩłå¼ł

0.28

rowsers

0.26

-utils

0.25

æİĸ

0.25

sav

0.25

éĴī

0.24

éĴĿ

0.24

TableName

0.24

é½¡

0.23

 precip

0.23

Activations Density 0.003%

Reg

No Comments

No Known Activations