INDEX

Explanations

repeated uses of the word "the"

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

 Mittelpunkt

-0.36

 conducta

-0.34

 resultat

-0.32

 frontières

-0.31

 inſ

-0.31

 собою

-0.30

 kartą

-0.29

 sociétés

-0.29

 собой

-0.29

 Erscheinung

-0.29

POSITIVE LOGITS

 estekak

0.80

 kaarangay

0.74

OGND

0.68

 phony

0.66

 متعلقه

0.65

principalColumn

0.65

 հղումներ

0.65

 pesky

0.65

WithIOException

0.64

󠁣

0.63

Activations Density 0.049%

repeated uses of the word "the"

No Comments

No Known Activations