INDEX

Explanations

-T

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.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

 Flour

-0.26

Ton

-0.24

-ton

-0.23

}}},↵

-0.23

æį¢æĿ¥

-0.23

edly

-0.23

ä¸Ģæĺ¯

-0.22

 Flux

-0.22

 INSTANCE

-0.22

ç¨Ģç¼º

-0.22

POSITIVE LOGITS

 after

0.70

 later

0.69

 during

0.69

 shortly

0.69

 soon

0.64

 afterward

0.62

 sooner

0.61

 when

0.56

 afterwards

0.55

ä¹ĭåĲİ

0.55

Activations Density 0.001%

-T

No Comments

No Known Activations