INDEX

Explanations

terms related to critical concepts and exploration, particularly in religious or scientific contexts

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

sum

-0.67

ImageIO

-0.59

netinet

-0.57

bbons

-0.56

HomeController

-0.55

-0.54

Heu

-0.53

ChildScrollView

-0.53

Bos

-0.52

Σ

-0.52

POSITIVE LOGITS

 citoy

0.89

 pauv

0.77

 kasarigan

0.77

 gjelder

0.75

 avoient

0.75

Critical

0.75

 barbati

0.74

 uſed

0.73

 stanga

0.73

 ſta

0.72

Activations Density 0.123%

terms related to critical concepts and exploration, particularly in religious or scientific contexts

No Comments

No Known Activations