INDEX

Explanations

patterns of numerical or coded information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

endphp

-0.77

fikasi

-0.70

 Buren

-0.66

 Processed

-0.62

angeran

-0.62

processed

-0.60

 rerum

-0.59

UrlEncoded

-0.58

mages

-0.58

hyung

-0.57

POSITIVE LOGITS

AndEndTag

0.87

OGND

0.71

ating

0.67

gggg

0.67

izing

0.65

Solving

0.60

ofire

0.60

UnusedPrivate

0.60

forChild

0.57

Taking

0.57

Activations Density 0.684%

patterns of numerical or coded information

No Comments

No Known Activations