INDEX

Explanations

discussions or mentions of future topics or sections to be addressed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 للاسماء

-0.77

featureID

-0.73

 وتسجيلات

-0.70

 autorytatywna

-0.69

PreferredItem

-0.66

ConstraintMaker

-0.66

 estekak

-0.63

CloseOperation

-0.63

 estimés

-0.62

出版年

-0.61

POSITIVE LOGITS

 discussed

1.17

discussed

1.15

 discuss

1.07

 discussion

1.03

 later

1.00

 briefly

0.97

 discussing

0.93

 hereinafter

0.92

 discusses

0.91

 DISCUSS

0.90

Activations Density 0.825%

discussions or mentions of future topics or sections to be addressed

No Comments

No Known Activations