INDEX

Explanations

terms and concepts related to scientific and medical classifications or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Komunikasi

-0.45

 telefónica

-0.41

หมด

-0.40

 działania

-0.39

 diinginkan

-0.39

 Realität

-0.39

 bancaria

-0.39

 Handlung

-0.39

 réunis

-0.39

 merawat

-0.39

POSITIVE LOGITS

tamol

0.56

saraba

0.54

 lenker

0.54

GEBURTSDATUM

0.52

Personensuche

0.51

ScopeManager

0.50

 ostavi

0.49

+#+#

0.49

 يتيمه

0.47

примеча

0.47

Activations Density 1.143%

terms and concepts related to scientific and medical classifications or conditions

No Comments

No Known Activations