INDEX

Explanations

phrases that are typically used as bylines or captions in articles or social media posts

oai_token-act-pair · gpt-3.5-turbo

instances of slashes used in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Beir

-0.78

acters

-0.77

 explan

-0.73

terday

-0.72

 defe

-0.70

 misunder

-0.70

ylum

-0.69

abies

-0.69

 disapp

-0.68

etime

-0.67

POSITIVE LOGITS

ËĪ

1.50

usr

0.94

Via

0.88

etc

0.84

wcsstore

0.81

Film

0.80

taboola

0.76

 Shutterstock

0.73

 caption

0.72

join

0.72

Activations Density 0.016%

phrases that are typically used as bylines or captions in articles or social media posts

instances of slashes used in a document

No Comments

No Known Activations