INDEX

Explanations

phrases related to family relationships and connections

oai_token-act-pair · gpt-3.5-turbo

references to family relationships, particularly maternal and paternal connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ioxide

-0.69

 surface

-0.67

IUM

-0.64

eport

-0.63

Vote

-0.62

OVER

-0.62

itivity

-0.62

itiveness

-0.61

ocations

-0.61

ohm

-0.61

POSITIVE LOGITS

thood

0.91

 grandparents

0.84

daughter

0.82

 daughters

0.80

 siblings

0.79

 daughter

0.78

 grandmother

0.76

iblings

0.76

 sister

0.75

 sons

0.74

Activations Density 0.317%

phrases related to family relationships and connections

references to family relationships, particularly maternal and paternal connections

No Comments

No Known Activations