Skillget

genpark-prompt-token-length-budget-packager-skill

mcp
Guvenlik Denetimi
Uyari
Health Uyari
  • No license — Repository has no license file
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 7 GitHub stars
Code Gecti
  • Code scan — Scanned 6 files during light audit, no dangerous patterns found
Permissions Gecti
  • Permissions — No dangerous permissions requested

Bu listing icin henuz AI raporu yok.

SUMMARY

Bin-packing optimizer grouping variable-length training examples into fixed context window token budgets

README.md

genpark-prompt-token-length-budget-packager-skill

A First-Fit Decreasing (FFD) bin packing optimizer designed to maximize context window utilization during LLM fine-tuning and batch inference.

Architecture

flowchart TD
    Examples[Variable Length Prompts] --> Estimator[Token Length Estimator]
    Estimator --> Sorter[Descending Length Sorter]
    Sorter --> FFD[First-Fit Decreasing Packer]
    FFD --> Bins[Optimized Fixed-Budget Context Bins]

Features

  • FFD Bin Packing: Minimizes padding waste across fixed token windows (2k, 4k, 8k, 32k).
  • Utilization Analytics: Calculates utilization rate and token waste per bin.
  • Pure Python Standard Library: No external dependencies.

Yorumlar (0)

Sonuc bulunamadi