INDEX

Explanations

references to terrorist attacks

oai_token-act-pair · gpt-3.5-turbo

references to terrorist attacks and related law enforcement activities

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wcsstore

-0.95

HUD

-0.85

 Navajo

-0.83

pod

-0.82

natureconservancy

-0.81

 Cherokee

-0.80

UGC

-0.77

DOE

-0.76

paio

-0.76

 Painter

-0.75

POSITIVE LOGITS

 attackers

1.43

 attacker

1.41

 gunmen

1.40

 Hebdo

1.38

 assailants

1.36

 jihadist

1.31

 assailant

1.30

 Tunis

1.25

 attacks

1.25

 terrorist

1.24

Activations Density 0.398%

references to terrorist attacks

references to terrorist attacks and related law enforcement activities

No Comments

No Known Activations