INDEX

Explanations

names of individuals or locations

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names or individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Arkham

-0.72

 Normandy

-0.71

 parole

-0.70

 Hitchcock

-0.69

 msec

-0.66

 Phillies

-0.65

 replay

-0.65

 NIGHT

-0.65

 BART

-0.64

AFL

-0.64

POSITIVE LOGITS

 Yuan

1.40

Xia

1.37

jiang

1.36

wei

1.34

 Xiang

1.33

 Jing

1.32

Cao

1.27

 Guang

1.26

 Ning

1.25

Bei

1.25

Activations Density 0.078%

names of individuals or locations

mentions of specific names or individuals

No Comments

No Known Activations