INDEX

Explanations

textual patterns related to scientific measurements

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 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

Vidite

-0.95

EndGlobalSection

-0.91



-0.80

Personendaten

-0.78

Aiheesta

-0.76

elemField

-0.76

 AspNetCore

-0.70

 ModelExpression

-0.70

Kariera

-0.70

/*++

-0.70

POSITIVE LOGITS

 snippetHide

0.64

popcorn

0.50

 plafond

0.49

))->

0.49

 débat

0.46

 bénévol

0.45

 popcorn

0.45

]?.

0.45

 fact

0.45

 Others

0.44

Activations Density 1.109%

textual patterns related to scientific measurements

No Comments

No Known Activations