INDEX

Explanations

of

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

äºº

-0.26

.by

-0.26

çİ»çĴĥ

-0.25

äººæµģ

-0.25

 overrun

-0.25

.cl

-0.24

æ¥¼ä¸ĭ

-0.24

gel

-0.24

 surrounds

-0.24

izr

-0.23

POSITIVE LOGITS

third

0.34

 fourth

0.33

 thirds

0.32

-quarters

0.27

ä¸īåĪĨä¹ĭä¸Ģ

0.27

 third

0.27

 eighth

0.26

peare

0.25

rias

0.25

 fours

0.25

Activations Density 0.009%

of

No Comments

No Known Activations