INDEX

Explanations

texts related to hoaxes and pranks

oai_token-act-pair · gpt-3.5-turbo

references to hoaxes and pranks

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

oyal

-0.77

foreseen

-0.71

avid

-0.69

HCR

-0.68

arching

-0.68

ailable

-0.68

apsed

-0.67

 obligated

-0.66

affer

-0.65

incinn

-0.64

POSITIVE LOGITS

 hoax

1.14

sters

0.95

ulence

0.84

ishly

0.84

 spoof

0.82

mong

0.78

erella

0.76

ument

0.76

 prank

0.75

gee

0.74

Activations Density 0.009%

texts related to hoaxes and pranks

references to hoaxes and pranks

No Comments

No Known Activations