training guide

Start and monitor compatible local training

Launch a user-configured toolkit and track only the process states the companion can observe.

Cracked-gray Curious Obmil carries a glowing device toward a smaller, sorrowful GoldCap with one broad main cap, exactly two smaller caps, and living gold veins in a device-local workshop glade
Curious brings the work home. GoldCap keeps the training path on the user’s own ground.
PREREQUISITES
  • Reviewed local dataset
  • Compatible base model and toolkit
  • Enough memory and storage
  • A tested command profile
SUCCESS EVIDENCE
  • A real local process started
  • Logs and output location are available
  • Completion is not confused with activation or quality
ORDERED PROCEDURE

Complete the task

01

Run the toolkit’s own preflight

02

Choose the reviewed dataset and base model

03

Start training explicitly

04

Watch process state, logs, progress fields, and output path

05

Pause, resume, or stop only when the toolkit supports it

06

Verify adapter files after completion

07

Test the selected runtime separately

08

Activate an adapter manually only after validation

Warnings
  • Training can fail after preparation succeeds
  • DecaCap cannot guarantee every model/toolkit pairing