INDEX

Explanations

words related to giving credit or attribution

oai_token-act-pair · gpt-3.5-turbo

references to sources or attribution in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

²¾

-0.74

 retard

-0.73

pup

-0.71

 Lans

-0.70

·

-0.68

rency

-0.66

velt

-0.65

estern

-0.64

improve

-0.64

Ñĭ

-0.64

POSITIVE LOGITS

Credit

1.08

 Credit

1.03

 Cards

0.83

 CARD

0.82

 Credits

0.79

 credit

0.79

enhagen

0.78

 Observatory

0.78

 Transfer

0.76

credit

0.76

Activations Density 0.008%

words related to giving credit or attribution

references to sources or attribution in a text

No Comments

No Known Activations