INDEX

Explanations

references to prestigious awards and honors, especially the Nobel Prize and the Pulitzer Prize

oai_token-act-pair · gpt-3.5-turbo

references to prestigious awards such as the Nobel and Pulitzer Prizes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Pep

-0.67

town

-0.67

nu

-0.66

 repe

-0.65

DEP

-0.65

ãĥ¯ãĥ³

-0.64

 frust

-0.64

URA

-0.64

cry

-0.63

gel

-0.63

POSITIVE LOGITS

 laureate

1.03

 Prize

0.99

 awarded

0.89

 Laure

0.77

 Writers

0.75

 scholar

0.73

 physicist

0.72

 Winner

0.72

 Journalists

0.72

 Institution

0.72

Activations Density 0.032%

references to prestigious awards and honors, especially the Nobel Prize and the Pulitzer Prize

references to prestigious awards such as the Nobel and Pulitzer Prizes

No Comments

No Known Activations