INDEX

Explanations

computer code or medical 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

Not in Any Lists

No Comments

Negative Logits

 regular

-0.28

ä¿¡åı·

-0.27

ì»µ

-0.27

 signals

-0.27

 Glover

-0.26

 Lyrics

-0.26

idge

-0.25

Ø¨ÙĦ

-0.25

 Fairfax

-0.25

DCALL

-0.25

POSITIVE LOGITS

éĶĢ

0.28

vect

0.28

å»¶æľŁ

0.27

 vect

0.27

sheet

0.26

enh

0.25

_<

0.25

 handleChange

0.25

OnChange

0.25

-Za

0.25

Activations Density 0.022%

computer code or medical text

No Comments

No Known Activations