INDEX

Explanations

mentions of the British royalty, specifically queens

oai_token-act-pair · gpt-3.5-turbo

mentions of "Queen" and related entities or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ramid

-0.77

anwhile

-0.69

herer

-0.69

nir

-0.67

enegger

-0.66

quickShipAvailable

-0.65

ython

-0.65

odcast

-0.65

sych

-0.63

vernment

-0.63

POSITIVE LOGITS

 Anne

1.15

 Elizabeth

1.07

 Mary

0.99

ston

0.99

stown

0.98

 Victoria

0.98

pin

0.96

 Alexandra

0.93

 Mother

0.92

Bee

0.88

Activations Density 0.016%

mentions of the British royalty, specifically queens

mentions of "Queen" and related entities or titles

No Comments

No Known Activations