INDEX

Explanations

phrases and words that are comparable or analogous in nature

oai_token-act-pair · gpt-3.5-turbo

references to comparisons and similarities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

erity

-1.03

Published

-0.81

 onwards

-0.73

ebook

-0.70

 furthermore

-0.67

Cause

-0.65

idable

-0.64

duction

-0.62

ITION

-0.61

confidence

-0.60

POSITIVE LOGITS

lihood

1.01

lier

0.78

 ours

0.73

how

0.72

 Watergate

0.68

 medieval

0.66

soDeliveryDate

0.64

 those

0.63

 what

0.62

 Schr

0.60

Activations Density 0.146%

phrases and words that are comparable or analogous in nature

references to comparisons and similarities

No Comments

No Known Activations