INDEX

Explanations

details about a specific item or object in a technical context, possibly related to production or assembly processes

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to specific car models and their historical significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 Flavoring

-0.99

iotic

-0.79

iotics

-0.78

EEK

-0.77

MN

-0.75

atonin

-0.74

 PLoS

-0.73

adelphia

-0.73

NPR

-0.72

izon

-0.72

POSITIVE LOGITS

 Ferrari

1.52

 Corvette

1.49

 Porsche

1.46

 Fiat

1.35

 automobiles

1.34

 cars

1.32

 Mercedes

1.31

 McLaren

1.31

 Bentley

1.30

borgh

1.30

Activations Density 0.422%

details about a specific item or object in a technical context, possibly related to production or assembly processes

references to specific car models and their historical significance

No Comments

No Known Activations