INDEX

Explanations

references to the Confederate era or the Civil War

oai_token-act-pair · gpt-3.5-turbo

references to the Confederate flag and related historical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•Embedding Space Digit Features

No Comments

Negative Logits

 Jordan

-0.72

 rehab

-0.70

 Schwartz

-0.69

 Shiny

-0.69

hart

-0.66

Kos

-0.66

HR

-0.66

stein

-0.66

sche

-0.66

 Rubin

-0.65

POSITIVE LOGITS

2.74

2.50

2.48

2.47

2.45

2.36

2.25

2.02

1.94

1.90

Activations Density 0.035%

references to the Confederate era or the Civil War

references to the Confederate flag and related historical context

No Comments

No Known Activations