INDEX

Explanations

words related to online activism, surveillance, and conspiracy theories

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

olen

-0.73

 Barron

-0.71

beit

-0.67

mounted

-0.66

ĳ

-0.64

arger

-0.62

SN

-0.62

nos

-0.62

reditary

-0.61

ī

-0.60

POSITIVE LOGITS

 spree

1.29

 regimen

1.20

 regime

1.13

 scheme

1.07

 programme

1.06

 debacle

1.03

 program

1.01

 affair

1.00

 apparatus

0.98

 fiasco

0.97

Activations Density 0.359%

words related to online activism, surveillance, and conspiracy theories

No Comments

No Known Activations

words related to online activism, surveillance, and conspiracy theories

No Comments

No Known Activations