INDEX

Explanations

proper nouns related to different people named Walter

oai_token-act-pair · gpt-3.5-turbo

the name "Walter" in various contexts

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

ebook

-1.02

arian

-0.87

ateur

-0.82

arians

-0.80

atories

-0.79

ional

-0.78

alities

-0.74

atory

-0.73

unity

-0.72

amia

-0.72

POSITIVE LOGITS

 Mond

0.95

 Walter

0.89

 Reed

0.87

 Winc

0.83

\\\\\\\\

0.75

 Payton

0.72

 Cron

0.72

 Isaac

0.72

fur

0.71

 Benjamin

0.71

Activations Density 0.005%

proper nouns related to different people named Walter

the name "Walter" in various contexts

No Comments

No Known Activations