verdict-coreml / verification-L512.json
alexwengg's picture
Bind Verdict parity reports to published Core ML package hashes
835208c verified
Raw History Blame Contribute Delete
7.71 kB
{
"source_repo": "heman10x/rlcd-modernbert-151m",
"source_revision": "8af2496eb63c7fa66d7d234e1f62629380030eb4",
"package": "verdict_fp16_L512_candidates25.mlpackage",
"package_files_sha256": {
"Data/com.apple.CoreML/model.mlmodel": "7ad21a9ee93d1c573e6d89e6c35511f08c984a5c47108bb2b908c96344c03903",
"Data/com.apple.CoreML/weights/weight.bin": "fb438de5823dd91c8448b373d026b7e6dcb7149bf871fde54057fd6d82586886",
"Manifest.json": "047cc80f7d66d822ca0d9c0100a6a6dc96ae2d70eac42e292a2b246e01e64ddd"
},
"chip": "arm",
"macos": "27.0",
"max_calibrated_probability_error": 0.0006975581321209678,
"selected_id_agreements": 5,
"questions": 5,
"rows": [
{
"name": "lost card",
"type": "choice",
"tokens": 53,
"candidates_including_abstention": 3,
"native": {
"type": "choice",
"selected_id": "card_lost",
"selected_probability": 0.6279954438939314,
"probabilities": {
"card_lost": 0.6279954438939314,
"pin_reset": 0.12699384187218735,
"__insufficient_evidence__": 0.2450107142338813
},
"concentration": 0.18185852825435922,
"is_abstention": false,
"p_abstain": 0.2450107142338813,
"calibration_status": "calibrated_for_scope",
"choice": "card_lost"
},
"coreml": {
"type": "choice",
"selected_id": "card_lost",
"selected_probability": 0.6286024758124755,
"probabilities": {
"card_lost": 0.6286024758124755,
"pin_reset": 0.12674703377209953,
"__insufficient_evidence__": 0.24465049041542497
},
"concentration": 0.18252696123986079,
"is_abstention": false,
"p_abstain": 0.24465049041542497,
"calibration_status": "calibrated_for_scope",
"choice": "card_lost"
},
"max_calibrated_probability_error": 0.0006070319185441653,
"selected_id_agrees": true,
"median_model_call_ms": 8.427584019955248
},
{
"name": "binary inquiry",
"type": "noul",
"tokens": 50,
"candidates_including_abstention": 3,
"native": {
"type": "noul",
"selected_id": "true",
"selected_probability": 0.4692844463560151,
"probabilities": {
"true": 0.4692844463560151,
"false": 0.24040823180168494,
"__insufficient_evidence__": 0.29030732184229996
},
"concentration": 0.03808302501474403,
"is_abstention": false,
"p_abstain": 0.29030732184229996,
"calibration_status": "calibrated_for_scope",
"noul": 0.6612502295701238,
"selected_outcome": "true"
},
"coreml": {
"type": "noul",
"selected_id": "true",
"selected_probability": 0.46998200448813604,
"probabilities": {
"true": 0.46998200448813604,
"false": 0.23993786932412314,
"__insufficient_evidence__": 0.29008012618774076
},
"concentration": 0.03846968937399986,
"is_abstention": false,
"p_abstain": 0.29008012618774076,
"calibration_status": "calibrated_for_scope",
"noul": 0.6620211967927305,
"selected_outcome": "true"
},
"max_calibrated_probability_error": 0.0006975581321209678,
"selected_id_agrees": true,
"median_model_call_ms": 8.437459007836878
},
{
"name": "urgency score",
"type": "score",
"tokens": 55,
"candidates_including_abstention": 4,
"native": {
"type": "score",
"selected_id": "__insufficient_evidence__",
"selected_probability": 0.3259148157325998,
"probabilities": {
"low": 0.19793214504236917,
"medium": 0.2898236019444062,
"high": 0.18632943728062487,
"__insufficient_evidence__": 0.3259148157325998
},
"concentration": 0.02039165010923949,
"is_abstention": true,
"p_abstain": 0.3259148157325998,
"calibration_status": "calibrated_for_scope",
"score": null,
"selected_level_id": "__insufficient_evidence__",
"selected_value": null
},
"coreml": {
"type": "score",
"selected_id": "__insufficient_evidence__",
"selected_probability": 0.3262236353210236,
"probabilities": {
"low": 0.19780841262370594,
"medium": 0.28950115689828004,
"high": 0.18646679515699038,
"__insufficient_evidence__": 0.3262236353210236
},
"concentration": 0.020408360566200767,
"is_abstention": true,
"p_abstain": 0.3262236353210236,
"calibration_status": "calibrated_for_scope",
"score": null,
"selected_level_id": "__insufficient_evidence__",
"selected_value": null
},
"max_calibrated_probability_error": 0.000322445046126163,
"selected_id_agrees": true,
"median_model_call_ms": 8.381415973417461
},
{
"name": "insufficient evidence",
"type": "choice",
"tokens": 41,
"candidates_including_abstention": 3,
"native": {
"type": "choice",
"selected_id": "__insufficient_evidence__",
"selected_probability": 0.6003396635578521,
"probabilities": {
"visa": 0.19581437244279193,
"mastercard": 0.2038459639993559,
"__insufficient_evidence__": 0.6003396635578521
},
"concentration": 0.1354398243565177,
"is_abstention": true,
"p_abstain": 0.6003396635578521,
"calibration_status": "calibrated_for_scope",
"choice": "__insufficient_evidence__"
},
"coreml": {
"type": "choice",
"selected_id": "__insufficient_evidence__",
"selected_probability": 0.6008594846026971,
"probabilities": {
"visa": 0.19567826624893642,
"mastercard": 0.2034622491483665,
"__insufficient_evidence__": 0.6008594846026971
},
"concentration": 0.13595645758378905,
"is_abstention": true,
"p_abstain": 0.6008594846026971,
"calibration_status": "calibrated_for_scope",
"choice": "__insufficient_evidence__"
},
"max_calibrated_probability_error": 0.000519821044845048,
"selected_id_agrees": true,
"median_model_call_ms": 8.738874981645495
},
{
"name": "BFCL-tool-selection:simple:simple_254",
"type": "choice",
"tokens": 283,
"candidates_including_abstention": 3,
"native": {
"type": "choice",
"selected_id": "yes",
"selected_probability": 0.5095718104586857,
"probabilities": {
"yes": 0.5095718104586857,
"no": 0.28447377613128705,
"__insufficient_evidence__": 0.20595441341002724
},
"concentration": 0.0655574261710149,
"is_abstention": false,
"p_abstain": 0.20595441341002724,
"calibration_status": "calibrated_for_scope",
"choice": "yes"
},
"coreml": {
"type": "choice",
"selected_id": "yes",
"selected_probability": 0.5093345700124705,
"probabilities": {
"yes": 0.5093345700124705,
"no": 0.28465867339056805,
"__insufficient_evidence__": 0.20600675659696147
},
"concentration": 0.06541626747983909,
"is_abstention": false,
"p_abstain": 0.20600675659696147,
"calibration_status": "calibrated_for_scope",
"choice": "yes"
},
"max_calibrated_probability_error": 0.00023724044621520335,
"selected_id_agrees": true,
"median_model_call_ms": 10.042167035862803
}
],
"passed": true
}