INDEX

Explanations

phrases indicating features or capabilities of products or services

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

<?

-0.45

Lähteet

-0.38

Witam

-0.35

CloseOperation

-0.33

 sukienka

-0.32

atouille

-0.31

glement

-0.30

 restauration

-0.29

psack

-0.29

wiście

-0.29

POSITIVE LOGITS

tagHelperRunner

0.76

󠁴

0.69

 propOrder

0.62

enterOuterAlt

0.60

NameInMap

0.60

 للاسماء

0.56

reportWebVitals

0.55

httphttps

0.52

orical

0.52

enderror

0.51

Activations Density 0.042%

phrases indicating features or capabilities of products or services

No Comments

No Known Activations