INDEX

Explanations

references to working conditions and the working class

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

 tambi

-0.58

 abstrait

-0.57

 volando

-0.56

bbene

-0.54

mű

-0.53

ktır

-0.52

 colpa

-0.52

 BIBSYS

-0.51

winners

-0.51

 pregunto

-0.50

POSITIVE LOGITS

 propOrder

0.92

 jScrollPane

0.80

 Working

0.70

Working

0.69

XmlEnum

0.68

 working

0.66

DeleteBehavior

0.65

Personendaten

0.64

AsUp

0.64

izing

0.63

Activations Density 0.105%

references to working conditions and the working class

No Comments

No Known Activations