INDEX

Explanations

prepositions, particularly 'in' and 'of' occurring near particular nouns

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

Query

-0.76

DragonMagazine

-0.74

 Cosponsors

-0.72

-0.68

-0.66

ãĤ³

-0.63

ãĥĩ

-0.63

Emb

-0.63

-0.62

POSITIVE LOGITS

 existence

1.19

 history

1.03

 terms

1.01

clusions

0.93

efficiency

0.92

 America

0.86

herent

0.83

 Europe

0.83

ducers

0.83

 humankind

0.82

Activations Density 0.213%

prepositions, particularly 'in' and 'of' occurring near particular nouns

No Comments

No Known Activations