INDEX

Explanations

emotion-related words, such as "excitement," "sadness," and "arrogance."

oai_token-act-pair · gpt-3.5-turbo

emotions and negative experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

REE

-0.65

Activity

-0.64

ppel

-0.63

onal

-0.62

ãĥīãĥ©

-0.61

LAND

-0.61

Go

-0.61

asketball

-0.60

WD

-0.59

ĵ

-0.59

POSITIVE LOGITS

iest

1.25

 inherent

1.11

iness

0.94

 emanating

0.94

 afforded

0.94

 surrounding

0.93

 plag

0.86

lessness

0.85

 quot

0.83

 aspect

0.82

Activations Density 0.296%

emotion-related words, such as "excitement," "sadness," and "arrogance."

emotions and negative experiences

No Comments

No Known Activations