INDEX

Explanations

references to specific chapters or sections within a larger document or book

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

berman

-0.72

zzle

-0.69

cffff

-0.68

pes

-0.68

oggles

-0.67

yg

-0.66

yre

-0.65

fitting

-0.64

enez

-0.64

wd

-0.62

POSITIVE LOGITS

ĸļ

1.06

 chapters

0.83

acters

0.83

naire

0.81

icularly

0.80

 chapter

0.79

 Chapters

0.79

hound

0.75

chapter

0.73

book

0.73

Activations Density 0.014%

references to specific chapters or sections within a larger document or book

No Comments

No Known Activations