INDEX

Explanations

spoilers in the text

oai_token-act-pair · gpt-3.5-turbo

references to spoilers in content

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

llan

-0.73

tis

-0.71

undai

-0.70

CV

-0.66

artney

-0.66

issance

-0.66

mu

-0.65

trak

-0.64

rys

-0.63

independent

-0.63

POSITIVE LOGITS

 spoilers

1.12

 spoiler

1.08

OIL

1.02

 spoil

0.99

 Ahead

0.93

 SECTION

0.86

 WARNING

0.85

 ahead

0.84

 Warning

0.83

 Spoiler

0.82

Activations Density 0.057%

spoilers in the text

references to spoilers in content

No Comments

No Known Activations