INDEX

Explanations

locations in various cities

oai_token-act-pair · gpt-3.5-turbo

locations and geographical references within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

selves

-0.76

matically

-0.71

onym

-0.66

ioned

-0.65

 beneficial

-0.64

 interaction

-0.64

rifice

-0.63

uristic

-0.62

 growth

-0.62

 advancements

-0.62

POSITIVE LOGITS

 Borough

0.89

 Calif

0.86

 District

0.86

 Sands

0.83

 Castle

0.79

 Plain

0.79

aka

0.77

 Meadows

0.76

 Seas

0.76

 whence

0.75

Activations Density 0.435%

locations in various cities

locations and geographical references within the text

No Comments

No Known Activations