INDEX

Explanations

Automatically generated text

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

cott

-0.27

rove

-0.26

abouts

-0.26

POINTS

-0.26

    ↵    ↵    ↵    ↵

-0.25

èµİ

-0.25

ç«

-0.25

ä¸ĢåĽŀäºĭ

-0.24

headed

-0.24

ÑģÐµÐ½

-0.24

POSITIVE LOGITS

 Natural

0.29

 naturally

0.28

èĩªçĦ¶

0.27

åĦĴ

0.27

 words

0.27

 natural

0.27

itra

0.27

Kle

0.26

natural

0.26

 dense

0.26

Activations Density 0.001%

Automatically generated text

No Known Activations