INDEX

Explanations

phrases indicating an increase or decline in something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating increases or decreases in various metrics or phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

soever

-0.70

lot

-0.69

pointers

-0.69

bryce

-0.65

 hooks

-0.65

ById

-0.64

IDE

-0.64

%%

-0.63

 pals

-0.63

 SERV

-0.62

POSITIVE LOGITS

efficiency

1.32

 regards

1.28

effic

1.27

 lieu

1.11

 relation

1.11

ordinate

1.09

 favor

1.08

 regard

1.07

clus

1.06

humane

1.06

Activations Density 0.507%

phrases indicating an increase or decline in something

phrases indicating increases or decreases in various metrics or phenomena

No Comments

No Known Activations