INDEX

Explanations

proper nouns and names, particularly related to individuals or entities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

expandindo

-0.51

astéroïdes

-0.43

piecze

-0.42

SourceChecksum

-0.42

findpost

-0.41

 kaarangay

-0.41

AndEndTag

-0.41

 препратки

-0.37

Termine

-0.36

 parcela

-0.36

POSITIVE LOGITS

 surla

0.57

'{@

0.55

 للمعارف

0.51

OGND

0.49

 فريبيس

0.48

 MatDialog

0.44

 '-';

0.42

 Wicidata

0.41

 newArray

0.40

 onBackPressed

0.40

Activations Density 0.055%

proper nouns and names, particularly related to individuals or entities

No Comments

No Known Activations