INDEX

Explanations

questions or statements expressing uncertainty or possibility

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty or questions about future events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

derived

-0.72

arers

-0.70

pecially

-0.70

APH

-0.69

advertisement

-0.69

 pione

-0.68

ustomed

-0.67

arest

-0.66

ascript

-0.66

ItemImage

-0.66

POSITIVE LOGITS

he

1.20

she

0.90

 Romo

0.88

 Dwight

0.86

 Jarrett

0.86

 Teddy

0.86

 Jimmy

0.86

 Trey

0.86

 they

0.84

AJ

0.83

Activations Density 0.392%

questions or statements expressing uncertainty or possibility

expressions of uncertainty or questions about future events

No Comments

No Known Activations