Files
clarify-rl-grpo-qwen3-1-7b-…/training_summary.json
ModelHub XC 192b4445a4 初始化项目,由ModelHub XC社区提供模型
Model: Kanan2005/clarify-rl-grpo-qwen3-1-7b-run6
Source: Original Platform
2026-07-17 13:09:12 +08:00

15 lines
625 B
JSON

{
"model": "Qwen/Qwen3-1.7B",
"max_steps": 300,
"num_generations": 8,
"vllm_gpu_memory_utilization": 0.55,
"max_completion_length": 1536,
"train_seconds": 4272.8014805316925,
"stats": "TrainOutput(global_step=300, training_loss=0.05546886101374791, metrics={'train_runtime': 4222.7288, 'train_samples_per_second': 0.568, 'train_steps_per_second': 0.071, 'total_flos': 0.0, 'train_loss': 0.05546886101374791})",
"failed": false,
"failure_reason": "",
"output_dir": "clarify-rl-grpo-qwen3-1-7b-run6",
"trackio_space_id": "clarify-rl-grpo-qwen3-1-7b-run6",
"num_log_entries": 301,
"smoke_test": false
}