INDEX

Explanations

references to templates and specific types of content related to them

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 AssemblyCulture

-0.65

parsedMessage

-0.52

 kasarigan

-0.43

xiety

-0.42

StoreMessageInfo

-0.39

 Umsatz

-0.39

 singuli

-0.39

 Fortsch

-0.38

 Profesional

-0.37

 Perſ

-0.37

POSITIVE LOGITS

Tem

1.99

tem

1.95

Tem

1.94

tem

1.63

TEM

1.59

TEM

1.57

 Temple

1.40

templ

1.38

Temple

1.35

Template

1.28

Activations Density 0.409%

references to templates and specific types of content related to them

No Known Activations