INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 constexpr

-0.51



-0.51

IsContent

-0.50

hringer

-0.48

 Continuum

-0.48

pousada

-0.47

étoit

-0.47

 stockbild

-0.46

 mCurrent

-0.46

 Majefty

-0.46

POSITIVE LOGITS

 stran

1.52

 resourceCulture

0.48

 jazy

0.43



0.43

stran

0.41

 lingue

0.40

 languages

0.38

 pages

0.36

 language

0.35

sidemargin

0.35

Activations Density 0.000%

No Known Activations

This feature has no known activations.