INDEX

Explanations

triggers and feeling

The neuron detects words and phrases signaling medical conditions, symptoms, risk factors, or treatment-related flare-ups.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 câncer

0.64

近代

0.63

癌症

0.60

 carcinomas

0.56

Inj

0.56

损伤

0.56

OJ

0.56

 인수

0.56

FindFileData

0.55

 Corre

0.54

POSITIVE LOGITS

 flare

1.39

 flares

1.31

flare

1.16

 triggers

1.14

 Flare

1.14

 episodes

1.09

 exacerb

1.09

 relapse

1.09

 flared

1.07

trigger

1.05

Activations Density 0.336%