INDEX

Explanations

mentions of the term "owl"

oai_token-act-pair · gpt-3.5-turbo

references to the character "Owl."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 fracturing

-0.69

 Palestin

-0.69

 PRESS

-0.66

 confir

-0.65

 hemor

-0.63

 Benz

-0.61

 fractured

-0.61

 epid

-0.60

senal

-0.60

 doctors

-0.59

POSITIVE LOGITS

owl

1.49

sburg

1.09

oon

1.03

sworth

0.95

oons

0.94

lihood

0.93

endor

0.91

ings

0.91

ows

0.91

osion

0.90

Activations Density 0.008%

mentions of the term "owl"

references to the character "Owl."

No Comments

No Known Activations