INDEX

Explanations

mentions of photography related words and activities

oai_token-act-pair · gpt-3.5-turbo

empty or placeholder segments in the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 incorpor

-0.69

_.

-0.65

jri

-0.65

eday

-0.64

 Vaugh

-0.63

 challeng

-0.62

enegger

-0.61

 disadvant

-0.58

atever

-0.58

 Niet

-0.57

POSITIVE LOGITS

Â¶

0.65

 Belfast

0.49

 Confederate

0.48

oplan

0.47

 âĢº

0.47

 Copyright

0.46

)]

0.46

zbollah

0.45

 Image

0.44

thood

0.43

Activations Density 1.835%

mentions of photography related words and activities

empty or placeholder segments in the text

No Comments

No Known Activations