INDEX

Explanations

images or information related to news or photography

oai_token-act-pair · gpt-3.5-turbo

references to Getty Images and related photo credit indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

yip

-0.85

jri

-0.71

 privilege

-0.68

 brill

-0.68

 unsus

-0.67

kered

-0.66

 infiltr

-0.66

luck

-0.64

 particip

-0.63

 unch

-0.63

POSITIVE LOGITS

 Images

2.16

sburg

1.81

Images

1.67

images

1.20

 Photos

1.05

Photos

1.05

 Features

1.04

 IMAGES

1.00

Image

1.00

 images

0.99

Activations Density 0.023%

images or information related to news or photography

references to Getty Images and related photo credit indicators

No Comments

No Known Activations