INDEX

Explanations

instances of something being officially recognized or approved after meeting certain standards

oai_token-act-pair · gpt-3.5-turbo

terms related to certification or official approval

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

————————

-0.71

OPLE

-0.69

leton

-0.69

ROR

-0.68

htar

-0.67

 Rove

-0.66

bill

-0.66

Alz

-0.64

âĿ

-0.63

hung

-0.63

POSITIVE LOGITS

 certification

1.26

 certified

1.21

 Certification

1.02

ificate

1.01

 Certified

0.94

 cert

0.89

 certify

0.87

cert

0.86

 certificate

0.77

rador

0.76

Activations Density 0.012%

instances of something being officially recognized or approved after meeting certain standards

terms related to certification or official approval

No Comments

No Known Activations