INDEX

Explanations

references to individuals named Nathan or Rick

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

Not in Any Lists

No Comments

Negative Logits

apiVersion

-0.81

LLocation

-0.79

 onCreateView

-0.75

Portale

-0.75

 TableColumn

-0.70

 Fuji

-0.67

(!__

-0.67

mybatisplus

-0.67

 Tanz

-0.65

={`/

-0.65

POSITIVE LOGITS

 Poker

0.88

 poker

0.87

Poker

0.77

poker

0.76

parsedMessage

0.75

Reno

0.73

 Tela

0.71

 Rhetoric

0.71

er

0.71

)});

0.71

Activations Density 0.025%

references to individuals named Nathan or Rick

No Comments

No Known Activations