INDEX

Explanations

references to quantity, particularly involving the number one and its variations in terms of counts and cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 Signalez

-0.73

Eup

-0.70

Datuak

-0.69

 Wiktionnaire

-0.69

thâu

-0.68

</tfoot>

-0.67

 impon

-0.66

 Sarm

-0.65

}');

-0.64

IntoConstraints

-0.64

POSITIVE LOGITS

 TextAppearance

0.50

two

0.50

 prior

0.49

 near

0.47

 noted

0.45

 excluded

0.44

 post

0.43

 στον

0.43

GraphicsUnit

0.42

had

0.42

Activations Density 0.623%

references to quantity, particularly involving the number one and its variations in terms of counts and cases

No Comments

No Known Activations