INDEX

Explanations

proper nouns related to political figures

oai_token-act-pair · gpt-3.5-turbo

references to Gaddafi and his roles or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

itialized

-0.78

Seattle

-0.77

Fra

-0.77

Colorado

-0.77

MW

-0.75

Madison

-0.73

Ago

-0.72

Indiana

-0.72

 Flavoring

-0.71

Brave

-0.70

POSITIVE LOGITS

addafi

1.47

 Gaddafi

1.29

 dictators

0.97

 dictatorship

0.94

 Tripoli

0.93

 regimes

0.93

ascus

0.89

 Hussein

0.89

 Libya

0.89

abad

0.88

Activations Density 0.011%

proper nouns related to political figures

references to Gaddafi and his roles or actions

No Comments

No Known Activations