INDEX

Explanations

references to law enforcement or police-related incidents

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

————

-0.80

module

-0.78

along

-0.74

 Proced

-0.73

asus

-0.68

anni

-0.67

anon

-0.66

idency

-0.66

emale

-0.66

qqa

-0.64

POSITIVE LOGITS

 slightest

0.97

 curs

0.86

 watered

0.81

 smallest

0.77

 slight

0.74

 superf

0.74

 modest

0.73

 mundane

0.71

 trivial

0.70

 occasional

0.69

Activations Density 0.922%

references to law enforcement or police-related incidents

No Comments

No Known Activations