INDEX

Explanations

references to San Francisco and its landmarks

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

);?>

-0.69

"}")

-0.68

 Pologne

-0.63

modelBuilder

-0.60

Jum

-0.59

prepareStatement

-0.59

 popular

-0.58

 Mortimer

-0.57

 Morley

-0.56

rzost

-0.56

POSITIVE LOGITS

 California

1.02

 CALIFORNIA

1.01

California

0.98

 Sonoma

0.94

 Berkeley

0.92

 Calif

0.91

 Oakland

0.90

Calif

0.90

 Francisco

0.89

 Menlo

0.87

Activations Density 0.097%

references to San Francisco and its landmarks

No Known Activations