INDEX

Explanations

specific terms related to entities and health-related concepts

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

 CreateTagHelper

-0.63

razor

-0.59

jooq

-0.57

RegisterType

-0.57

Edg

-0.56

parsedMessage

-0.56

 Púb

-0.54

 bày

-0.54

aita

-0.53

 kelahiran

-0.53

POSITIVE LOGITS

 Roses

0.69

 Parcel

0.69

 sportifs

0.64

 parcel

0.63

čne

0.59

 parcels

0.57

Parcel

0.57

atsby

0.56

berries

0.55

 Sudbury

0.55

Activations Density 0.048%

specific terms related to entities and health-related concepts

No Comments

No Known Activations