INDEX

Explanations

numeric values

oai_token-act-pair · gpt-3.5-turbo

references to statistics or data related to health and societal issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

sic

-0.65

esy

-0.61

chenko

-0.59

icter

-0.58

 manoeuv

-0.58

Pwr

-0.57

icons

-0.57

EStreamFrame

-0.57

Removed

-0.55

});

-0.55

POSITIVE LOGITS

 convened

0.64

 federally

0.61

 nationally

0.60

 Amtrak

0.60

NPR

0.59

 Garfield

0.57

 Hartford

0.57

 Kenneth

0.57

 headquartered

0.57

 divided

0.56

Activations Density 1.272%

numeric values

references to statistics or data related to health and societal issues

No Comments

No Known Activations