INDEX

Explanations

B

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ready

-0.25

ä¿¡è´·

-0.25

æĺİ

-0.25

indi

-0.25

 culture

-0.25

St

-0.24

åĩºåĽ½

-0.24

net

-0.24

åı£ä¸Ń

-0.24

 going

-0.23

POSITIVE LOGITS

 breathed

0.27

å¼ķãģį

0.26

éħįå¥Ĺ

0.25

yme

0.25

çıı

0.24

UserCode

0.24

/cms

0.24

å®ļçĤ¹

0.24

å¹¶å°Ĩ

0.23

ymes

0.23

Activations Density 0.002%

B

No Comments

No Known Activations