INDEX

Explanations

keywords related to importance or centrality

oai_token-act-pair · gpt-3.5-turbo

references to primary or central topics in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

omic

-0.69

ossession

-0.67

 Cron

-0.65

 Leone

-0.64

AUT

-0.63

 Kidd

-0.62

 Vaugh

-0.61

 Lynch

-0.60

 McKenna

-0.60

ï¸

-0.60

POSITIVE LOGITS

stay

1.41

tenance

1.04

 arteries

0.93

 artery

0.89

 culprit

0.86

yout

0.84

 proponent

0.82

lining

0.79

 sticking

0.78

tan

0.78

Activations Density 0.018%

keywords related to importance or centrality

references to primary or central topics in various contexts

No Comments

No Known Activations