INDEX

Explanations

phone numbers and addresses

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

ependent

-0.65

istically

-0.65

isexual

-0.64

ependence

-0.63

earchers

-0.60

bye

-0.59

 accessibility

-0.58

 behav

-0.57

itiz

-0.57

ument

-0.56

POSITIVE LOGITS

MSN

1.10

 Fahrenheit

0.70

 Rated

0.67

0.65

INS

0.64

Merit

0.63

 Oops

0.62

 SERV

0.62

 acres

0.61

0.60

Activations Density 0.222%

phone numbers and addresses

No Comments

No Known Activations