INDEX

Explanations

occurrences of the word "arrive" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 transfieras

-0.71

Portail

-0.69

Diwedd

-0.68

 Houſe

-0.68

 disambiguazione

-0.67

memoized

-0.67

 againſt

-0.64

verwijspagina

-0.63

rungsseite

-0.63

 osal

-0.62

POSITIVE LOGITS

 arrived

0.92

 arrives

0.88

 arrive

0.84

arrived

0.83

pp

0.78

 Arrived

0.74

 imagines

0.74

 imagined

0.73

 Arri

0.73

 imagine

0.70

Activations Density 0.065%

occurrences of the word "arrive" in various forms

No Comments

No Known Activations