INDEX

Explanations

phrases related to the payment platform "Paypal"

oai_token-act-pair · gpt-3.5-turbo

references to the word "pal" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

DEF

-0.78

DT

-0.75

 Tucker

-0.71

 Definitive

-0.65

 redef

-0.64

 Vacc

-0.63

 Bulldogs

-0.63

 Everett

-0.62

CPS

-0.62

 Discovery

-0.61

POSITIVE LOGITS

pal

4.60

Pal

2.48

pal

1.66

Pal

1.55

pak

1.16

ipal

1.14

PAL

1.07

pad

1.06

 palms

1.04

opal

1.04

Activations Density 0.008%

phrases related to the payment platform "Paypal"

references to the word "pal" in various contexts

No Comments

No Known Activations