INDEX

Explanations

locations or positions of people or objects

oai_token-act-pair · gpt-3.5-turbo

references to the state of being or location of subjects

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iously

-0.75

iencies

-0.71

inous

-0.70

 Consumer

-0.64

icals

-0.63

tsy

-0.62

 Jackets

-0.62

uthor

-0.60

 Cosby

-0.60

UTH

-0.59

POSITIVE LOGITS

 located

1.13

 headquartered

1.07

 weakest

1.06

 headed

1.05

 situated

1.03

 stationed

1.02

 positioned

0.97

 heading

0.95

 aiming

0.93

 concerned

0.89

Activations Density 0.106%

locations or positions of people or objects

references to the state of being or location of subjects

No Comments

No Known Activations