INDEX

Explanations

instances where something is upside down

oai_token-act-pair · gpt-3.5-turbo

references to potential benefit or positive outcomes

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

ãģ®éŃĶ

-0.93

ãģ®å®

-0.87

Rush

-0.74

tein

-0.73

 Annotations

-0.71

atter

-0.67

assetsadobe

-0.66

ART

-0.66

ocal

-0.66

 Cosponsors

-0.66

POSITIVE LOGITS

 upside

1.66

 srfAttach

0.88

 sacrific

0.83

 ceiling

0.79

 horizont

0.79

 perspect

0.78

 downside

0.78

olicy

0.77

 handc

0.76

 surv

0.75

Activations Density 0.006%

instances where something is upside down

references to potential benefit or positive outcomes

No Comments

No Known Activations