File size: 1,292 Bytes
b5cc210 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | {
"status": "fallback",
"backend": "env_reward_fallback",
"model_id": "Qwen/Qwen2.5-1.5B-Instruct",
"records": 1,
"prompts_path": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/data/processed/training_corpus_grpo_prompts.jsonl",
"reward_summary": {
"count": 1,
"avg_reward": 0.764,
"avg_reward_components": {
"format_compliance_score": 0.999,
"candidate_alignment_score": 0.999,
"legality_score": 0.999,
"safety_delta_score": 0.5,
"burden_improvement_score": 0.5,
"disease_stability_score": 0.9,
"dosing_quality_score": 0.5,
"abstention_quality_score": 0.56,
"efficiency_score": 0.857,
"process_fidelity_score": 0.92,
"explanation_grounding_score": 0.8,
"anti_cheat_score": 0.001,
"uncertainty_calibration_score": 0.7
},
"avg_primary_reward_channels": {
"safety_legality": 0.675,
"clinical_improvement": 0.633,
"dosing_quality": 0.53,
"process_integrity": 0.894
}
},
"reward_log": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/checkpoints/grpo_reward_components.jsonl",
"train_metrics": {
"steps_executed": 1.0
},
"artifact_path": "",
"unsloth_available": false,
"trl_runtime_error": "forced_fallback"
} |