INDEX

Explanations

references to or mentions of terrorists

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "terrorists" and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Ecc

-0.73

Ã°

-0.72

ilde

-0.72

ld

-0.70

 Ã¾

-0.68

 Craft

-0.67

yd

-0.67

icka

-0.66

EE

-0.64

ithing

-0.63

POSITIVE LOGITS

 terrorists

3.49

 extremists

2.22

 jihadists

2.16

 terrorist

2.06

 terrorism

2.02

 militants

1.99

 criminals

1.94

terror

1.90

 Islamists

1.88

 insurgents

1.88

Activations Density 0.021%

references to or mentions of terrorists

occurrences of the word "terrorists" and related terms

No Comments

No Known Activations