INDEX

Explanations

names of people or entities that are being accused of something

oai_token-act-pair · gpt-3.5-turbo

names of individuals or entities involved in accusations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

æĸ¹

-0.87

antha

-0.71

ONSORED

-0.71

 attest

-0.63

 largeDownload

-0.62

thus

-0.62

ourney

-0.62

areth

-0.61

redo

-0.60

arious

-0.59

POSITIVE LOGITS

 unfairly

0.86

 wrongly

0.73

 unjust

0.71

 failing

0.68

 foul

0.67

of

0.66

 inaction

0.66

 falsely

0.65

cheat

0.65

tics

0.65

Activations Density 0.106%

names of people or entities that are being accused of something

names of individuals or entities involved in accusations

No Comments

No Known Activations