INDEX

Explanations

statements related to conspiracy theories and political controversies

oai_token-act-pair · gpt-3.5-turbo

references to claims and accusations related to Barack Obama and Donald Trump

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Pwr

-0.83

cms

-0.81

 partName

-0.75

NAD

-0.73

 Indust

-0.71

byss

-0.70

 overload

-0.69

 Despair

-0.67

CSI

-0.67

 sqor

-0.66

POSITIVE LOGITS

 wiret

1.06

 pard

0.97

 presidency

0.94

 presidential

0.93

 impeachment

0.92

 aides

0.89

 assassinated

0.89

0.88

 impe

0.88

 pardon

0.88

Activations Density 0.430%

statements related to conspiracy theories and political controversies

references to claims and accusations related to Barack Obama and Donald Trump

No Comments

No Known Activations