INDEX

Explanations

steel

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

çĲ¥

-0.34

Stone

-0.29

 Pompe

-0.28

æľ¨è´¨

-0.27

åĩºçīĪ

-0.27

 rubble

-0.26

éĹ¾

-0.26

èĥ¶

-0.26

anted

-0.25

 Stone

-0.25

POSITIVE LOGITS

 steel

0.56

éĴ¢æĿĲ

0.55

éĴ¢æĿ¿

0.52

éĴ¢

0.52

ä¸įéĶĪéĴ¢

0.49

steel

0.49

åĲĪéĩĳ

0.49

 stainless

0.49

éĭ¼

0.48

 alloy

0.47

Activations Density 0.032%

steel

No Comments

No Known Activations