INDEX

Explanations

proper nouns or names, specifically 'Campbell'

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Campbell."

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

lihood

-0.90

olitics

-0.82

liest

-0.73

initialized

-0.72

peed

-0.70

LEASE

-0.70

liness

-0.68

eways

-0.67

fix

-0.66

ntil

-0.65

POSITIVE LOGITS

 Soup

1.03

 Campbell

0.95

otte

0.93

iac

0.87

ibur

0.80

shire

0.79

icum

0.78

 Newman

0.78

rade

0.77

gren

0.76

Activations Density 0.014%

proper nouns or names, specifically 'Campbell'

mentions of the name "Campbell."

No Comments

No Known Activations