INDEX

Explanations

mentions of misinformation, hoaxes, and pranks

oai_token-act-pair · gpt-3.5-turbo

terms related 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 @ 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

bar

-0.71

 Apply

-0.70

pec

-0.70

 guiActiveUnfocused

-0.67

Bar

-0.67

istered

-0.67

Apply

-0.67

bar

-0.67

 tender

-0.66

strength

-0.65

POSITIVE LOGITS

 hoax

3.94

 prank

2.00

 disinformation

1.71

 misinformation

1.64

 spoof

1.46

 falsehood

1.34

 faked

1.27

geries

1.27

 myths

1.21

 deception

1.18

Activations Density 0.041%

mentions of misinformation, hoaxes, and pranks

terms related to hoaxes and pranks

No Comments

No Known Activations