INDEX

Explanations

phrases that indicate roles or functions served by subjects within contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

migrationBuilder

-0.75

 Taktlose

-0.71

:✨

-0.67

 Sanger

-0.61

expandindo

-0.61

rungsseite

-0.61

cambrian

-0.60

 enjo

-0.57

 estimés

-0.57

InlineData

-0.57

POSITIVE LOGITS

 toimi

0.66

 fungerer

0.64

inghouse

0.64

 serve

0.63

ˏ

0.62

Shutterstock

0.61

起到

0.60

 consultato

0.60

 recueil

0.60

 används

0.60

Activations Density 0.205%

phrases that indicate roles or functions served by subjects within contexts

No Comments

No Known Activations