INDEX

Explanations

text

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

Cub

-0.28

æľĢçµĤ

-0.27

‘

-0.26

 Coch

-0.26

 final

-0.25

ics

-0.25

 gross

-0.24

 Chad

-0.24

uples

-0.24

ifiable

-0.24

POSITIVE LOGITS

CRT

0.26

[mid

0.25

adem

0.25

older

0.24

ãģĹãģ¦ãģıãģłãģķãģĦ

0.24

ç²¹

0.23

weg

0.23

owe

0.23

 yours

0.23

 theirs

0.23

Activations Density 23.219%

text

No Comments

No Known Activations