INDEX

Explanations

personal relationships and knowledge-related phrases

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

hement

-0.78

 phased

-0.68

ONSORED

-0.67

owered

-0.66

ģĸ

-0.65

 ratification

-0.63

IRO

-0.62

cation

-0.61

BuyableInstoreAndOnline

-0.60

imaru

-0.59

POSITIVE LOGITS

 intimately

1.05

 whereabouts

0.93

 firsthand

0.86

 beforehand

0.81

 secrets

0.78

 instinctively

0.75

 workings

0.74

ographies

0.73

 existed

0.72

 exists

0.71

Activations Density 0.244%

personal relationships and knowledge-related phrases

No Comments

No Known Activations