spark-training-gotchas

SKILLFlujo de trabajocomunidad
v0.0.0wshobsonMITActualizado hace 16 dFuente →

Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.

Community-submitted skill. Not yet reviewed by the Forge team. Full prompt content may not be available.Request review →
39kEstrellas del repo
1Clientes
1Formatos
hace 16 dÚltima actualización
Skill
Autorwshobson
Versión0.0.0
LicenciaMIT
CategoríaFlujo de trabajo
Formatosskill.md
PromptAbrir (ver la pestaña Prompt)
Compatibilidad
Claude✓ Compatible
Cursor—
Copilot—
ChatGPT—
Gemini—
Acerca de

Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.

Palabras clave
skillclaude