INDEX

Explanations

news-related terms and phrases

oai_token-act-pair · gpt-3.5-turbo

elements related to multimedia content and articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

,,,,

-0.57

adobe

-0.57

 cart

-0.54

sheets

-0.54

 somewhere

-0.54

 neglig

-0.53

packed

-0.52

naire

-0.52

 Astro

-0.52

 Infinite

-0.52

POSITIVE LOGITS

 WATCHED

0.87

zens

0.70

 Coverage

0.67

iban

0.65

 Continued

0.64

Reuters

0.64

oppers

0.63

 VIDEO

0.63

 REUTERS

0.62

resa

0.62

Activations Density 0.151%

news-related terms and phrases

elements related to multimedia content and articles

No Comments

No Known Activations