INDEX

Explanations

references to French culture, people, and language

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 ennemi

-0.64

 ainfi

-0.61

 bénéficiaire

-0.56

 feroit

-0.56

 gouver

-0.55

 démocr

-0.55

 neutre

-0.53

 fidé

-0.53

 enfans

-0.52

 Jérusalem

-0.52

POSITIVE LOGITS

riezmann

0.60

 LeBlanc

0.59

 المعيارى

0.54

 Dreyfus

0.51

niques

0.51

 Caron

0.51

 French

0.49

 Haiti

0.48

 Paris

0.47

 clich

0.47

Activations Density 0.401%

references to French culture, people, and language

No Comments

No Known Activations