INDEX

Explanations

words related to possessive relationships

oai_token-act-pair · gpt-3.5-turbo

references to the pronouns "your" and "Your"

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

trak

-0.80

apo

-0.79

 Appears

-0.75

ilts

-0.73

Lago

-0.70

raq

-0.68

rix

-0.68

 Goes

-0.68

aneers

-0.68

ega

-0.67

POSITIVE LOGITS

own

1.37

 favorite

1.02

 favourite

1.01

 mileage

0.96

 choice

0.89

selves

0.89

 guys

0.86

 opponent

0.86

 fingertips

0.85

 imagination

0.84

Activations Density 0.106%

words related to possessive relationships

references to the pronouns "your" and "Your"

No Comments

No Known Activations