INDEX

Explanations

references to distances or spatial relationships

oai_token-act-pair · gpt-3.5-turbo

phrases indicating small quantities or distances

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Reviewer

-0.67

 Recap

-0.65

udeb

-0.64

ushima

-0.63

abwe

-0.62

-+

-0.61

theme

-0.61

abeth

-0.60

 Excellence

-0.59

ufact

-0.59

POSITIVE LOGITS

 yards

1.67

 metres

1.64

 feet

1.61

 meters

1.59

 kilometers

1.41

 miles

1.40

 kilometres

1.40

 centimeters

1.37

feet

1.33

 inches

1.31

Activations Density 0.140%

references to distances or spatial relationships

phrases indicating small quantities or distances

No Comments

No Known Activations