INDEX

Explanations

phrases related to surveillance activities

oai_token-act-pair · gpt-3.5-turbo

references to surveillance practices and programs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

gran

-0.81

versions

-0.76

adding

-0.74

interstitial

-0.74

Join

-0.74

gs

-0.74

apply

-0.73

orian

-0.72

brance

-0.72

onement

-0.71

POSITIVE LOGITS

 surveillance

1.32

illance

1.23

 Surveillance

1.17

 cameras

0.94

ank

0.94

 scanners

0.80

 screenings

0.79

 systems

0.79

 Systems

0.78

 sweeps

0.77

Activations Density 0.011%

phrases related to surveillance activities

references to surveillance practices and programs

No Comments

No Known Activations