INDEX

Explanations

words related to acts of terrorism

oai_token-act-pair · gpt-3.5-turbo

references to terrorism and related activities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elf

-0.85

laus

-0.83

ership

-0.79

)=(

-0.77

galitarian

-0.75

shire

-0.71

NRS

-0.70

Lago

-0.69

 Wine

-0.68

ais

-0.67

POSITIVE LOGITS

 terrorist

1.11

 terrorists

0.99

 mastermind

0.97

 bombers

0.97

terrorist

0.94

 bombings

0.93

 bombing

0.93

abad

0.92

 bomber

0.90

 assassin

0.89

Activations Density 0.011%

words related to acts of terrorism

references to terrorism and related activities

No Comments

No Known Activations