INDEX

Explanations

references to horses and equine-related terms

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

MLLoader

-0.53

WireFormatLite

-0.50

Secrets

-0.48

 Secrets

-0.47

 Dijk

-0.47

 viscose

-0.47

########.

-0.47

چال

-0.47

 الرياضيه

-0.47

itola

-0.46

POSITIVE LOGITS

 horses

1.02

 caballos

0.99

Horses

0.98

 horse

0.97

 Horses

0.94

 Horse

0.94

 pony

0.91

 caballo

0.87

 chevaux

0.86

horse

0.85

Activations Density 0.066%

references to horses and equine-related terms

No Comments

No Known Activations