INDEX

Explanations

words related to central or key locations

oai_token-act-pair · gpt-3.5-turbo

references to various hubs or centers of activity

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

IGHTS

-0.81

DonaldTrump

-0.75

chnology

-0.73

essee

-0.73

 graded

-0.71

issance

-0.67

ittee

-0.67

 extingu

-0.66

efe

-0.66

apons

-0.66

POSITIVE LOGITS

bub

1.43

bard

1.13

hub

1.10

hub

1.07

 hubs

1.01

lins

0.95

ris

0.94

lot

0.90

Hub

0.89

bies

0.89

Activations Density 0.017%

words related to central or key locations

references to various hubs or centers of activity

No Comments

No Known Activations