INDEX

Explanations

phrases indicating relationships and interactions among people or entities

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

Халык

-0.48

InSection

-0.41

 surla

-0.41

 nebude

-0.36

Predecesor

-0.36

 Memor

-0.32

 vorhanden

-0.32

 SIMBAD

-0.32

pholes

-0.31

 bibit

-0.31

POSITIVE LOGITS

 ويكيميديا

0.55

theoremstyle

0.51

sendStatus

0.49

resave

0.49

seros

0.48

 betweenstory

0.48

 artikkeli

0.48

niająca

0.48

 defaultstate

0.47

fromLTRB

0.47

Activations Density 0.735%

phrases indicating relationships and interactions among people or entities

No Comments

No Known Activations