INDEX

Explanations

arded

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.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

VERT

-0.28

 themselves

-0.27

 after

-0.26

ä¹ħ

-0.26

 rocks

-0.26

 already

-0.26

 itself

-0.25

 vert

-0.25

 unreachable

-0.25

 recognizable

-0.24

POSITIVE LOGITS

 constituents

0.31

 constitu

0.30

loor

0.29

 Constit

0.26

 constit

0.25

 constituent

0.25

 lucky

0.25

 constituency

0.25

*č↵

0.24

 draining

0.24

Activations Density 0.019%

arded

No Comments

No Known Activations