INDEX

Explanations

phrases indicating relationships or associations with other elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/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.10.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

 Syrie

-0.45

tagHelperRunner

-0.42



-0.40

 aprendido

-0.39

 africain

-0.38

 Artículo

-0.38

 venezolano

-0.36

UIControlState

-0.36

 murió

-0.35

 térmico

-0.35

POSITIVE LOGITS

 AssemblyCulture

0.49

มี

0.49

fieras

0.47

Rhestr

0.47

 wobei

0.47

 otomatig

0.47

 waarbij

0.46

تفصیلات

0.46

fony

0.46

hubanes

0.45

Activations Density 0.424%

phrases indicating relationships or associations with other elements

No Comments

No Known Activations