INDEX

Explanations

medical-related acronyms and medical procedures

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

hift

-0.68

 Berm

-0.67

xual

-0.64

Hog

-0.63

 Destination

-0.63

olesterol

-0.62

terday

-0.62

 Spectrum

-0.62

 Shakespeare

-0.60

NAD

-0.60

POSITIVE LOGITS

alach

1.23

ointment

1.07

eared

0.99

earance

0.97

reciation

0.97

ortion

0.97

ropri

0.96

legate

0.95

arel

0.95

eas

0.95

Activations Density 1.888%

medical-related acronyms and medical procedures

No Comments

No Known Activations