INDEX

Explanations

political and governmental terms and actions

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

ahime

-0.83

anmar

-0.68

utenberg

-0.66

enhagen

-0.65

maxwell

-0.64

itone

-0.63

ĸļ

-0.63

anchester

-0.63

ilib

-0.61

pha

-0.61

POSITIVE LOGITS

 whatsoever

1.94

 imaginable

1.32

soever

1.25

 conceivable

1.01

 except

0.99

 thereof

0.95

 anywhere

0.85

 besides

0.85

 hereafter

0.83

 thereafter

0.82

Activations Density 0.219%

political and governmental terms and actions

No Comments

No Known Activations