INDEX

Explanations

references to the word "Globe" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

Noi

-0.68

 Sien

-0.62

 extrait

-0.62

 Réponses

-0.61

 seid

-0.61

 pecuni

-0.61

 Orchard

-0.60

 Allende

-0.60

 Beller

-0.59

findall

-0.59

POSITIVE LOGITS

Glo

1.73

GLO

1.71

glo

1.64

 Glover

1.55

Glo

1.52

glo

1.38

 Globes

1.35

 Globe

1.32

 globes

1.23

GLO

1.23

Activations Density 0.008%

references to the word "Globe" and its variations

No Comments

No Known Activations