INDEX

Explanations

references to awards and accolades in the context of achievements, particularly in the arts and entertainment sectors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Reſ

-1.01

 كومونز

-0.83

Personensuche

-0.81

 myſelf

-0.75

Eſ

-0.74

 esternos

-0.74

 Houſe

-0.73

 Inſ

-0.73

abestanden

-0.71

 Conſ

-0.70

POSITIVE LOGITS

 National

0.73

 World

0.66

 International

0.65

 Global

0.57

0.56

National

0.56

International

0.55

ISES

0.55

 American

0.54

 Euro

0.53

Activations Density 0.549%

references to awards and accolades in the context of achievements, particularly in the arts and entertainment sectors

No Comments

No Known Activations