INDEX

Explanations

references to relationships between people

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

essen

-0.78

 percentages

-0.74

 quantities

-0.64

 ILCS

-0.64

 doses

-0.61

AE

-0.61

 partName

-0.60

 combination

-0.59

Premium

-0.58

 explor

-0.58

POSITIVE LOGITS

 hers

1.64

 ours

1.59

 mine

1.41

 yours

1.37

 theirs

1.34

 sorts

1.07

 Mine

1.00

mire

0.79

Mine

0.75

our

0.70

Activations Density 0.082%

references to relationships between people

No Comments

No Known Activations