INDEX

Explanations

references to academic publications and research across various scientific disciplines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

anément

-0.43

 يتيمه

-0.43

retudo

-0.41

SharedDtor

-0.40

 Krakowie

-0.39

 Olímpicos

-0.38

 caído

-0.38

 ilustração

-0.38

 yalnızca

-0.37

 científicas

-0.37

POSITIVE LOGITS

 Lokales

0.54

Réalisation

0.49

ppuden

0.46

Conserv

0.46

abestanden

0.45

Kaieteur

0.45

 Mute

0.44

IsContent

0.44

Soe

0.43

igor

0.42

Activations Density 0.450%

references to academic publications and research across various scientific disciplines

No Comments

No Known Activations