| { | |
| "model": "/output/job_e2e_1774222987861/merged/trainer-e2e-job_e2e_1774222987861", | |
| "metrics": { | |
| "model": "/output/job_e2e_1774222987861/merged/trainer-e2e-job_e2e_1774222987861", | |
| "samples": 1, | |
| "parseSuccessRate": 0.0, | |
| "mae": null, | |
| "rmse": null, | |
| "exactRate": 0.0, | |
| "within1Rate": 0.0, | |
| "within2Rate": 0.0, | |
| "meanSignedError": null, | |
| "avgPredictedScore": null, | |
| "parseErrors": 1, | |
| "inferenceErrors": 1, | |
| "emptyResponses": 1 | |
| }, | |
| "rows": [ | |
| { | |
| "sampleId": "eval_1", | |
| "question": "Сколько будет 2+2?", | |
| "candidateAnswer": "4", | |
| "referenceScore": 5.0, | |
| "maxScore": 5, | |
| "hashTags": [ | |
| "math", | |
| "sanity" | |
| ], | |
| "predictedScore": null, | |
| "predictedFeedback": null, | |
| "rawResponse": "", | |
| "parseError": true, | |
| "inferenceError": true, | |
| "absoluteError": null, | |
| "error": "'Qwen2Attention' object has no attribute 'apply_qkv'" | |
| } | |
| ] | |
| } |