INDEX

Explanations

mentions of Shanghai and related institutions

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 Moab

-0.43

 Canberra

-0.43

 Lehigh

-0.39

vermögen

-0.33

 Newt

-0.33

leg

-0.32

ever

-0.32

 florales

-0.32

 समीक्षक

-0.31

 Gettysburg

-0.31

POSITIVE LOGITS

 Shanghai

0.89

 propOrder

0.86

Shanghai

0.86

 Milan

0.82

Milan

0.77

 shanghai

0.76

 Mumbai

0.76

 Bombay

0.74

 Karachi

0.71

Mumbai

0.70

Activations Density 0.011%

mentions of Shanghai and related institutions

No Comments

No Known Activations