INDEX

Explanations

locations or organizations

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

 Notting

-0.85

framework

-0.72

metic

-0.70

amins

-0.68

 Material

-0.68

ãĤ©

-0.66

lihood

-0.65

feeding

-0.65

wagen

-0.64

hold

-0.63

POSITIVE LOGITS

IA

1.22

BI

1.22

OTUS

1.19

UL

1.16

RA

1.14

INT

1.13

FR

1.12

OST

1.12

VE

1.11

VEN

1.11

Activations Density 0.052%

locations or organizations

No Comments

No Known Activations