INDEX

Explanations

phrases related to attempting or experimenting with something

oai_token-act-pair · gpt-3.5-turbo

intentions or suggestions to attempt actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

head

-0.73

dylib

-0.66

oola

-0.64

hip

-0.63

thus

-0.63

DragonMagazine

-0.62

lav

-0.62

edly

-0.62

\-

-0.60

Newsletter

-0.60

POSITIVE LOGITS

 unsuccessfully

1.02

nir

0.81

 harder

0.77

ichick

0.74

amins

0.68

ipers

0.67

 desperately

0.67

ange

0.66

anke

0.65

onies

0.65

Activations Density 0.047%

phrases related to attempting or experimenting with something

intentions or suggestions to attempt actions

No Comments

No Known Activations