INDEX

Explanations

locations and geographical references in narratives

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

 Etrus

-0.83

Personendaten

-0.80

 eiffel

-0.75

providedIn

-0.73

 Crimean

-0.72

Бахар

-0.70

 Eiffel

-0.69

 Euphrates

-0.69

 Florentine

-0.68

 Algerian

-0.68

POSITIVE LOGITS

 central

1.01

 northern

0.94

 southern

0.94

 western

0.93

 rural

0.90

 eastern

0.86

 downtown

0.85

 modern

0.83

 northwestern

0.79

the

0.76

Activations Density 0.446%

locations and geographical references in narratives

No Comments

No Known Activations