INDEX

Explanations

rhyme

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enary

-0.28

ities

-0.28

zelf

-0.27

 Speakers

-0.26

ties

-0.25

engo

-0.25

ropri

-0.25

uments

-0.25

 widest

-0.24

mobx

-0.24

POSITIVE LOGITS

du

0.28

 tune

0.25

 exists

0.25

åĳ¨è¾¹

0.25

 substitute

0.25

exists

0.25

US

0.24

—but

0.24

 aValue

0.23

æĮĤåľ¨

0.23

Activations Density 0.190%

rhyme

No Comments

No Known Activations