INDEX

Explanations

references to cartoons

oai_token-act-pair · gpt-3.5-turbo

references to cartoons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

changes

-0.80

rity

-0.73

ivation

-0.71

ulia

-0.69

 Availability

-0.69

 Priv

-0.68

work

-0.67

locks

-0.66

 Recovery

-0.65

Skill

-0.64

POSITIVE LOGITS

 cartoon

3.71

 cartoons

3.15

 Cartoon

2.29

 caric

2.05

 caricature

1.91

 satir

1.64

 comic

1.57

 comics

1.44

 satirical

1.42

 animated

1.32

Activations Density 0.021%

references to cartoons

references to cartoons

No Comments

No Known Activations