INDEX

Explanations

proper nouns representing individuals or organizations

oai_token-act-pair · gpt-3.5-turbo

references to a specific entity or organization

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

glers

-0.83

GGGGGGGG

-0.74

ahime

-0.73

 Knights

-0.67

 Firefly

-0.65

phal

-0.65

BuyableInstoreAndOnline

-0.64

istically

-0.63

 Mali

-0.62

 Brist

-0.62

POSITIVE LOGITS

orted

1.42

rint

1.25

orters

1.22

orter

1.19

utation

1.18

utable

1.15

ository

1.14

utations

1.13

ublic

1.12

ositories

1.07

Activations Density 0.010%

proper nouns representing individuals or organizations

references to a specific entity or organization

No Comments

No Known Activations