INDEX

Explanations

addresses and locations, especially related to a specific landmark

oai_token-act-pair · gpt-3.5-turbo

phrases indicating significant public or social reactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Â·Â·

-0.87

apo

-0.83

pdf

-0.73

arations

-0.68

SPONSORED

-0.67

erity

-0.66

uphem

-0.64

"—

-0.64

 aver

-0.64

 thereby

-0.64

POSITIVE LOGITS

 popularity

1.23

 hype

0.99

 trend

0.96

 fanbase

0.91

 renown

0.88

 rave

0.86

 notoriety

0.85

 resurgence

0.82

 fame

0.80

 popular

0.80

Activations Density 0.552%

addresses and locations, especially related to a specific landmark

phrases indicating significant public or social reactions

No Comments

No Known Activations