INDEX

Explanations

phrases related to the use of force, particularly excessive or deadly force

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

itte

-0.86

algia

-0.84

alg

-0.84

hof

-0.77

hai

-0.74

stellar

-0.71

chart

-0.71

gres

-0.71

idges

-0.69

landish

-0.69

POSITIVE LOGITS

 unarmed

1.14

 unlawfully

1.03

 violence

0.97

 intimidation

0.97

 killings

0.94

 assault

0.91

 confrontation

0.91

 indiscrim

0.91

 tactics

0.89

 brutality

0.88

Activations Density 0.186%

phrases related to the use of force, particularly excessive or deadly force

No Comments

No Known Activations