INDEX

Explanations

specific mentions of vampires

oai_token-act-pair · gpt-3.5-turbo

references to vampires and vampire-related media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

giving

-0.78

beh

-0.77

placed

-0.74

essler

-0.73

ear

-0.73

abis

-0.71

orative

-0.70

olulu

-0.69

rontal

-0.69

ional

-0.68

POSITIVE LOGITS

 Vampire

1.19

 vampires

1.05

 vampire

1.01

ampire

0.98

 Dracula

0.95

 Masquerade

0.94

 squid

0.93

wolves

0.83

ampires

0.83

 Slayer

0.82

Activations Density 0.009%

specific mentions of vampires

references to vampires and vampire-related media

No Comments

No Known Activations