INDEX

Explanations

phrases indicating a lack of something

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate deficiencies or omissions in information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

arte

-0.86

 helicop

-0.74

Republic

-0.71

 Volunte

-0.66

bane

-0.66

uits

-0.65

Aid

-0.65

Europe

-0.65

Mobile

-0.64

assetsadobe

-0.64

POSITIVE LOGITS

 emphasis

1.07

 flaws

1.06

 similarities

1.02

 complexity

1.02

 twists

1.01

 clauses

1.00

 elements

0.97

 resemblance

0.95

 caveats

0.94

 similarity

0.93

Activations Density 0.689%

phrases indicating a lack of something

phrases that indicate deficiencies or omissions in information

No Comments

No Known Activations