INDEX

Explanations

IEEE

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 traction

-0.26

territ

-0.25

ByExample

-0.25

 genomic

-0.25

åĬĽ

-0.25

åĮĸçĸĹ

-0.25

 territory

-0.25

æĹ¶éĹ´åĨħ

-0.24

BUY

-0.24

_mono

-0.23

POSITIVE LOGITS

rams

0.28

Not

0.28

DMI

0.27

Not

0.27

NOT

0.26

åħļæĢ»

0.25

æ³¡

0.25

åīĶ

0.25

ãĥĶãĥ¼

0.25

àµº

0.25

Activations Density 20.190%

IEEE

No Comments

No Known Activations