INDEX

Explanations

words related to strong negative sentiments or emotions, particularly hatred

oai_token-act-pair · gpt-3.5-turbo

expressions of dislike or hatred toward various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

DragonMagazine

-1.08

igmatic

-0.93

ItemImage

-0.89

OGR

-0.83

aunder

-0.82

enture

-0.78

arov

-0.78

eva

-0.77

akeru

-0.76

aqu

-0.75

POSITIVE LOGITS

fully

1.03

 hated

0.96

 wasting

0.88

 hate

0.86

 Mondays

0.85

lessly

0.83

hate

0.82

 hates

0.77

 dearly

0.76

 bullies

0.72

Activations Density 0.061%

words related to strong negative sentiments or emotions, particularly hatred

expressions of dislike or hatred toward various subjects

No Comments

No Known Activations