INDEX

Explanations

knowledge

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

éĻį

-0.29

é¥µ

-0.27

esso

-0.27

ä¸įæħİ

-0.25

uraa

-0.25

/format

-0.25

ardless

-0.25

æĿ¥çļĦ

-0.24

ä¸Ģçľ¼

-0.24

azing

-0.24

POSITIVE LOGITS

 understanding

0.32

 knowledge

0.30

 understand

0.28

å¥ł

0.28

æĸĩæĺĮ

0.26

 Understanding

0.26

NES

0.25

äºĨè§£

0.25

çļĦçĲĨè§£

0.24

çĲĨè§£

0.24

Activations Density 0.028%

knowledge

No Comments

No Known Activations