INDEX

Explanations

spoiler-related content within text

oai_token-act-pair · gpt-3.5-turbo

terms related to spoilers in narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ients

-0.71

des

-0.70

ald

-0.67

 meters

-0.66

Ã

-0.66

 cent

-0.66

iking

-0.65

kn

-0.65

 calls

-0.65

 initiatives

-0.65

POSITIVE LOGITS

 spoiler

3.98

 Spoiler

3.55

Spoiler

3.06

 spoilers

2.64

oiler

2.58

OIL

2.22

 spoil

1.53

 disclaimer

1.42

 teaser

1.41

Warning

1.39

Activations Density 0.045%

spoiler-related content within text

terms related to spoilers in narratives

No Comments

No Known Activations