INDEX

Explanations

expressions related to negative traits or behaviors, particularly selfishness

oai_token-act-pair · gpt-3.5-turbo

concepts and discussions surrounding selfishness and altruism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Downloadha

-1.05

BuyableInstoreAndOnline

-0.77

annis

-0.75

gran

-0.74

UN

-0.69

AUT

-0.68

ORN

-0.68

enegger

-0.66

 Room

-0.66

ANN

-0.64

POSITIVE LOGITS

istical

1.07

istic

1.06

 minded

0.97

istically

0.92

 altru

0.88

 selfish

0.86

 motive

0.84

rifice

0.84

ly

0.83

bies

0.81

Activations Density 0.017%

expressions related to negative traits or behaviors, particularly selfishness

concepts and discussions surrounding selfishness and altruism

No Comments

No Known Activations