INDEX

Explanations

numbered lists or bullet points in a document

oai_token-act-pair · gpt-3.5-turbo

patterns of punctuation and spacing 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

 estranged

-0.78

 synthes

-0.75

 hust

-0.71

 endeav

-0.71

 hiber

-0.69

 freel

-0.69

ĸļ

-0.65

 Stard

-0.65

soever

-0.65

 wedd

-0.65

POSITIVE LOGITS

Notable

1.11

Reader

1.00

Bonus

0.96

Advertisement

0.93

Trivia

0.92

Favorite

0.91

Conclusion

0.89

âĵĺ

0.89

Score

0.88

?????-?????-

0.87

Activations Density 0.351%

numbered lists or bullet points in a document

patterns of punctuation and spacing in the text

No Comments

No Known Activations