INDEX

Explanations

Red

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

-0.41

inals

-0.28

ities

-0.28

stores

-0.27

ystals

-0.26

ngen

-0.26

servers

-0.26

 Penny

-0.25

ships

-0.25

åľ¨éĩĮéĿ¢

-0.25

POSITIVE LOGITS

0.45

ei

0.39

een

0.39

0.38

ing

0.38

ection

0.38

eed

0.35

ele

0.35

yz

0.34

Activations Density 0.269%

Red

No Comments

No Known Activations