INDEX

Explanations

words related to unauthorized access and device security

oai_token-act-pair · gpt-3.5-turbo

terms related to unauthorized access and implications of wearable technology

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

George

-0.92

 George

-0.80

aturdays

-0.79

 Napoleon

-0.76

Rush

-0.75

 vetoed

-0.75

xon

-0.75

 Slaughter

-0.74

Anth

-0.73

rican

-0.73

POSITIVE LOGITS

IoT

2.04

 wearable

1.84

 Bluetooth

1.79

 sensors

1.73

 devices

1.68

 wireless

1.56

 Devices

1.54

 gadgets

1.51

 device

1.50

 sensor

1.48

Activations Density 0.400%

words related to unauthorized access and device security

terms related to unauthorized access and implications of wearable technology

No Comments

No Known Activations