INDEX

Explanations

phrases related to ensuring safety, security, and responsibility

oai_token-act-pair · gpt-3.5-turbo

phrases related to security and safety

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

milo

-0.84

hops

-0.77

ibu

-0.77

orius

-0.76

ecided

-0.75

arrow

-0.74

culated

-0.71

elman

-0.71

otom

-0.70

quickShipAvailable

-0.70

POSITIVE LOGITS

 wellbeing

1.66

 preservation

1.55

 sake

1.47

 safety

1.44

 survival

1.31

 continuation

1.26

 enjoyment

1.26

 advancement

1.25

 stability

1.24

 welfare

1.22

Activations Density 0.216%

phrases related to ensuring safety, security, and responsibility

phrases related to security and safety

No Comments

No Known Activations