INDEX

Explanations

references to people and their attributes or actions in a conversational context

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 Taktlose

-0.52

 فريبيس

-0.48

Personensuche

-0.42

expandindo

-0.34

 pronto

-0.33

!(:

-0.33

 zichzelf

-0.33

 forêt

-0.32

 AssemblyCompany

-0.31

 Stages

-0.31

POSITIVE LOGITS

orianCalendar

0.53

domada

0.47

RRect

0.47

 المعيارى

0.45

Portail

0.43

vernote

0.43

 transférez

0.42

setObjectName

0.42

GeneratedValue

0.41

ULF

0.41

Activations Density 0.072%

references to people and their attributes or actions in a conversational context

No Comments

No Known Activations