INDEX

Explanations

references to specific organizations and geographic locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 gebou

-0.75

 gewicht

-0.73

 moeite

-0.72

 redenen

-0.71

 voeten

-0.70

 apparaat

-0.68

 verbonden

-0.64

 handen

-0.63

 punten

-0.62

 betaal

-0.61

POSITIVE LOGITS

 Dutch

0.83

van

0.83

 Maastricht

0.77

ngdoc

0.77

)['

0.76

abestanden

0.75

 Janssen

0.75

 Nieuw

0.74

 Eindhoven

0.74

enoord

0.73

Activations Density 0.216%

references to specific organizations and geographic locations

No Comments

No Known Activations