INDEX

Explanations

phrases indicating quantity or duration of time

oai_token-act-pair · gpt-3.5-turbo

phrases indicating quantities, particularly multiples of two

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hess

-0.72

bryce

-0.71

 Bomber

-0.67

Reviewer

-0.65

↵Âł

-0.64

Cole

-0.63

Clar

-0.63

 Havana

-0.63

 Cassidy

-0.62

TC

-0.62

POSITIVE LOGITS

 hundred

1.09

 thousand

1.09

 dozen

1.07

 million

0.89

 thirds

0.88

dozen

0.87

 billion

0.78

ousands

0.78

 inches

0.74

fold

0.73

Activations Density 0.132%

phrases indicating quantity or duration of time

phrases indicating quantities, particularly multiples of two

No Comments

No Known Activations