nick7nlp/Awesome-LLM-On-Policy-Distillation — GitHub trend

◂ BACK TO LEADERBOARD

FINDING #1585 · UNIT ID 1201758422

nick7nlp/Awesome-LLM-On-Policy-Distillation

A curated collection of papers and resources on On-Policy Distillation for Large Language Models.

[ PYTHON ][ GITHUB ↗ ]

SURPRISE SCORE

0.00

Score Breakdown

SURPRISE0.0375

ENGAGEMENT0.0717

FRESHNESS1.37

SCORE = SURPRISE × ENGAGEMENT^0.7 × FRESHNESS × VISIBILITY × CONFIDENCE

SURPRISE = WINDOW STARS / DAYS / (AUDIENCE + FLOOR)

29% OF STARS IN ARCHIVE

Growth Telemetry

VELOCITY /D

4.33

ACCEL

-0.08

RETENTION

34.7%

PEAK 2026-06-30 · FORK-RETENTION 0.0% · 130 STARS / WINDOW

Author Audience

AUDIENCE

FOLLOWERS

OWNER ★

607

Engagement Signals

FORKS

ISSUE AUTH

PR AUTH

UNIQUE STARGAZERS 130 / 130 (DIVERSITY 1.00)

Why This Is A Finding

nick7nlp/Awesome-LLM-On-Policy-Distillation собрал 130 звёзд за окно, тогда как у автора всего 15 подписчиков — эффективная аудитория ≈ 76. Это даёт surprise-индекс 0.0375 (звёзды относительно охвата автора, а не в абсолюте). Удержание форков 0.0% и 0 внешних контрибьюторов отделяют реальный инструмент от разовой вспышки. Акселерация отрицательная — внимание остывает после пика.

Related Findings

RANKS ABOVE 75% OF 6295 FINDINGS

#01shy3130/tickflow-stock-panel[ PYTHON ]0.46 #06Panniantong/Agent-Reach[ PYTHON ]0.29 #10sums001/Windows-Copilot-API[ PYTHON ]0.27 #12HUANGCHIHHUNGLeo/claude-real-video[ PYTHON ]0.24 #14realiti4/claude-swap[ PYTHON ]0.21

METRICS IN CONTEXT

MEDIAN ACROSS ALL 6295 FINDINGS · Δ vs MEDIAN · PERCENTILE = SHARE RANKED BELOW

METRICVALUEMEDIANΔ MEDPERCENTILE

SCORE0.010.00+0.01ABOVE 75%

VELOCITY4.334.23+0.10ABOVE 51%

RETENTION34.7%29.4%+5.3 PPABOVE 60%

FORKS892-84ABOVE 7%

SURPRISE0.040.01+0.03ABOVE 81%