INDEX

Explanations

instances of accusations or attributions of actions to individuals or entities

oai_token-act-pair · gpt-3.5-turbo

accusations related to wrongdoing or misconduct

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

etheless

-0.84

along

-0.70

erm

-0.69

 awaits

-0.67

 unfolds

-0.67

isSpecialOrderable

-0.67

fortunately

-0.65

 thereof

-0.65

req

-0.64

izons

-0.63

POSITIVE LOGITS

illet

0.83

 racist

0.79

 violent

0.75

 improper

0.75

BuyableInstoreAndOnline

0.73

 wiret

0.69

 incendiary

0.69

 Rahman

0.67

 collaborators

0.66

 cannibal

0.66

Activations Density 0.282%

instances of accusations or attributions of actions to individuals or entities

accusations related to wrongdoing or misconduct

No Comments

No Known Activations