初始化项目,由ModelHub XC社区提供模型

Model: promotion/ronpo-qwen3-8b-fair-sppo-avg-s42
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-07-18 13:23:11 +08:00
commit 9315ced0ff
18 changed files with 4248 additions and 0 deletions

20
training_status.json Normal file
View File

@@ -0,0 +1,20 @@
{
"status": "completed",
"candidate_id": "sppo_b",
"method": "sppo_avg",
"seed": 42,
"gpu": 3,
"pid": 762647,
"config": "/NHNHOME/26msit001_A/BASE/AIPR/sjkim/revision_qwen3_8b/full_iter1/fair_demo_20260715/sweep/candidates/sppo_b/config.yaml",
"dataset": "/NHNHOME/26msit001_A/BASE/AIPR/sjkim/revision_qwen3_8b/full_iter1/flagship_20260712/precomputed/avg",
"optimizer_steps": 900,
"effective_batch_size": 16,
"wandb_run_id": "a2bd72efe9d9",
"wandb_url": "https://wandb.ai/promotion-kim/mnpo/runs/a2bd72efe9d9",
"started_at": "2026-07-15T13:38:53+09:00",
"returncode": 0,
"model_complete": true,
"finite_results": true,
"measured_step": null,
"completed_at": "2026-07-15T15:40:43+09:00"
}