INDEX

Explanations

mentions of different types of residences

oai_token-act-pair · gpt-3.5-turbo

instances of the word "residence" and related terms

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

romeda

-0.83

anson

-0.71

displayText

-0.68

reg

-0.68

umbs

-0.67

Cola

-0.66

soDeliveryDate

-0.66

sid

-0.65

aneously

-0.65

essen

-0.65

POSITIVE LOGITS

 halls

0.93

 Halls

0.88

maid

0.88

chool

0.85

 residence

0.83

llah

0.80

loo

0.78

cation

0.73

velt

0.70

iture

0.69

Activations Density 0.019%

mentions of different types of residences

instances of the word "residence" and related terms

No Comments

No Known Activations