INDEX

Explanations

Code and licenses

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

Whilst

-0.67

 Whilst

-0.63

 recognised

-0.59

 whilst

-0.58

 personalised

-0.56

 postcode

-0.55

 realised

-0.53

 authorised

-0.53

 Maths

-0.52

Colour

-0.52

POSITIVE LOGITS

 federally

0.42

 supplemental

0.39

 downtown

0.38

Michigan

0.36

 Michigan

0.36

 Oklahoma

0.36

AVOR

0.36

 Illinois

0.35

Minnesota

0.35

avors

0.35

Activations Density 0.005%

Code and licenses

No Comments

No Known Activations