INDEX

Explanations

images in a document

oai_token-act-pair · gpt-3.5-turbo

mentions of images and visual content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

#$

-0.72

verning

-0.70

pai

-0.70

TRY

-0.70

cffff

-0.68

termination

-0.68

lie

-0.63

asonable

-0.63

oshi

-0.63

ority

-0.62

POSITIVE LOGITS

 images

3.61

Images

2.51

 image

2.48

 Images

2.47

 photographs

2.46

 pictures

2.38

images

2.38

 photos

2.35

 imagery

2.32

image

2.04

Activations Density 0.020%

images in a document

mentions of images and visual content

No Comments

No Known Activations