INDEX

Explanations

words related to international conflicts and relations, particularly regarding Iran, Iraq, and Syria

oai_token-act-pair · gpt-3.5-turbo

references to geopolitical issues involving Iran and Iraq

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ATA

-0.69

}}}

-0.67

document

-0.67

 Quadro

-0.67

 Hitchcock

-0.66

é¾įåĸļå£«

-0.65

erick

-0.64

nw

-0.64

 Guinness

-0.64

nant

-0.63

POSITIVE LOGITS

 uranium

0.92

 nuclear

0.82

ghazi

0.81

 consulate

0.79

 dictator

0.76

 favorably

0.76

 threats

0.75

 retaliation

0.74

 Iran

0.73

 plutonium

0.73

Activations Density 0.150%

words related to international conflicts and relations, particularly regarding Iran, Iraq, and Syria

references to geopolitical issues involving Iran and Iraq

No Comments

No Known Activations