INDEX

Explanations

1

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

 pres

-0.30

 centrif

-0.28

chet

-0.28

 PRES

-0.27

 Crawford

-0.27

ç²¾å½©

-0.27

Cos

-0.26

 pris

-0.26

 chin

-0.26

arda

-0.26

POSITIVE LOGITS

éĺŁ

0.29

ĽĦ

0.26

 Brave

0.26

Ð±Ð¾Ð»

0.25

è½µ

0.24

nik

0.24

è½¬è¿Ĳ

0.24

é¢ĦæĸĻ

0.24

 downhill

0.23

getInstance

0.23

Activations Density 0.000%

1

No Comments

No Known Activations