INDEX

Explanations

relevant to social or political controversies that provoke strong emotional reactions

oai_token-act-pair · gpt-3.5-turbo

expressions of public outrage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Solitaire

-0.72

Password

-0.70

wagen

-0.69

ramer

-0.68

properties

-0.67

ession

-0.67

uchin

-0.67

 cleaners

-0.67

 cleaner

-0.66

changed

-0.65

POSITIVE LOGITS

ously

1.26

iously

1.02

 indignation

0.96

 uproar

0.81

èĥ

0.76

ingly

0.75

uously

0.75

 outcry

0.73

 loudly

0.73

 outrage

0.72

Activations Density 0.026%

relevant to social or political controversies that provoke strong emotional reactions

expressions of public outrage

No Comments

No Known Activations