INDEX

Explanations

words related to family relationships, specifically focusing on spouses or partners

oai_token-act-pair · gpt-3.5-turbo

mentions of wives or the concept of marriage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

orescent

-0.82

umbn

-0.76

anyl

-0.72

obyl

-0.70

GoldMagikarp

-0.67

 constitu

-0.66

 inoc

-0.66

aucuses

-0.66

ococ

-0.66

omin

-0.65

POSITIVE LOGITS

wife

1.20

hood

0.96

 Maria

0.81

 Wife

0.80

 wife

0.80

friend

0.77

Maria

0.76

husband

0.76

gdala

0.75

doctor

0.73

Activations Density 0.016%

words related to family relationships, specifically focusing on spouses or partners

mentions of wives or the concept of marriage

No Comments

No Known Activations