INDEX

Explanations

instances of the word "Both" followed by a comparison between two entities or topics

oai_token-act-pair · gpt-3.5-turbo

references to the term "Both" indicating a comparison or duality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ugu

-0.76

heat

-0.73

itated

-0.71

cycl

-0.67

plete

-0.66

atur

-0.66

 horizon

-0.64

gin

-0.63

lÃ©

-0.63

ocratic

-0.63

POSITIVE LOGITS

 sexes

1.00

 genders

0.92

 sides

0.88

Both

0.78

 halves

0.76

wcs

0.74

theless

0.73

 formats

0.73

terness

0.70

 Ends

0.70

Activations Density 0.010%

instances of the word "Both" followed by a comparison between two entities or topics

references to the term "Both" indicating a comparison or duality

No Comments

No Known Activations