INDEX

Explanations

instances of the word "in" at the start of sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.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

 faveur

-0.52

 disambiguazione

-0.52

Personendaten

-0.51

Personensuche

-0.44

Transkript

-0.44

 للاسماء

-0.44

OGND

-0.43

 المعيارى

-0.43

Portail

-0.43

évaluateur

-0.42

POSITIVE LOGITS

 addition

1.59

 fact

1.45

 particular

1.07

fact

0.97

 tillegg

0.96

 turn

0.95

addition

0.94

 essence

0.89

 contrast

0.84

 reality

0.78

Activations Density 0.368%

instances of the word "in" at the start of sentences

No Comments

No Known Activations