INDEX

Explanations

phrases related to a specific studio, especially one named "Studio 9."

oai_token-act-pair · gpt-3.5-turbo

references to specific studios and their projects

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

ought

-0.72

ezvous

-0.72

 refere

-0.67

theless

-0.67

pend

-0.66

*/(

-0.66

IGH

-0.65

 occurrences

-0.65

owship

-0.64

cies

-0.64

POSITIVE LOGITS

 Studio

0.87

Lens

0.80

rait

0.77

Works

0.76

Actor

0.75

Studio

0.75

 Painter

0.73

 Suite

0.72

Comm

0.71

 Syndicate

0.71

Activations Density 0.012%

phrases related to a specific studio, especially one named "Studio 9."

references to specific studios and their projects

No Comments

No Known Activations