INDEX

Explanations

worth/value

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åĬ¨çī©

-0.30

æĢ»

-0.29

ogle

-0.27

dis

-0.26

 dominating

-0.26

ilt

-0.26

 guts

-0.26

ially

-0.24

 fais

-0.24

 observ

-0.24

POSITIVE LOGITS

=value

0.38

ä»·åĢ¼

0.35

alue

0.34

çļĦä»·åĢ¼

0.33

åĥ¹åĢ¼

0.31

/value

0.30

.Value

0.30

å°Ħæīĭ

0.29

ä¾¡

0.29

=Value

0.28

Activations Density 0.004%

worth/value

No Comments

No Known Activations