INDEX

Explanations

references to names and specific characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Administrativna

-0.76

 invokingState

-0.59

 للاسماء

-0.55

 صوتيه

-0.53

 Italijanski

-0.53

 שוליים

-0.52

Архівовано

-0.52

GEBURTSDATUM

-0.51

原始内容存档于

-0.51

ôtel

-0.51

POSITIVE LOGITS

 concentrating

0.34

 titles

0.33

 some

0.33

 instruments

0.32

 tools

0.31

 everything

0.31

 kõik

0.30

る

0.30

 følgende

0.30

'{@

0.29

Activations Density 0.415%

references to names and specific characters

No Comments

No Known Activations