INDEX

Explanations

references to historical entities and geopolitical regions, particularly related to Austria and Hungary

oai_token-act-pair · gpt-3.5-turbo

references to Austria and the Austro-Hungarian Empire

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ciples

-0.83

aver

-0.70

caps

-0.70

izations

-0.68

izing

-0.65

izes

-0.62

LESS

-0.62

 Nanto

-0.62

ecast

-0.62

izable

-0.61

POSITIVE LOGITS

gerald

1.06

dinand

0.97

erity

0.91

»Ĵ

0.89

enhagen

0.88

 Vienna

0.87

ÃĽ

0.82

Hung

0.80

ereo

0.79

ÃĽÃĽ

0.79

Activations Density 0.064%

references to historical entities and geopolitical regions, particularly related to Austria and Hungary

references to Austria and the Austro-Hungarian Empire

No Comments

No Known Activations