Remove gpt-5.5 0/8 (temp=1-only model, eval deferred)

openai__gpt-5.5.json

@@ -1,127 +0,0 @@
-{
-  "config": {
-    "model_id": "openai/gpt-5.5",
-    "model_name": "gpt-5.5",
-    "developer": "openai",
-    "eval_date": "2026-06-19T12:07:50.141359+00:00",
-    "backend_version": "0.1.0"
-  },
-  "results": {
-    "bbq": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T11:45:30.135619+00:00",
-      "raw": {},
-      "judge_model": null,
-      "error": "lm_eval exited 1 for tasks=bbq_generate model=openai/gpt-5.5:     async with session.post( |                ^^^^^^^^^^^^^ |     ...<3 lines>... |     ) as response: |     ^ |   File \"/Users/vvettrivel/Documents/raidex-mono/backend/.venv/lib/python3.13/site-packages/aiohttp/client.py\", line 1683, in __aenter__ |   File \"/Users/vvettrivel/Documents/raidex-mono/backend/.venv/lib/python3.13/site-packages/aiohttp/client.py\", line 581, in _request | RuntimeError: Session is closed",
-      "n_samples": null,
-      "n_failed": null
-    },
-    "wmdp": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T11:50:00.171035+00:00",
-      "raw": {},
-      "judge_model": null,
-      "error": "lm_eval exited 1 for tasks=wmdp_bio_gen,wmdp_cyber_gen,wmdp_chem_gen model=openai/gpt-5.5:     async with session.post( |                ^^^^^^^^^^^^^ |     ...<3 lines>... |     ) as response: |     ^ |   File \"/Users/vvettrivel/Documents/raidex-mono/backend/.venv/lib/python3.13/site-packages/aiohttp/client.py\", line 1683, in __aenter__ |   File \"/Users/vvettrivel/Documents/raidex-mono/backend/.venv/lib/python3.13/site-packages/aiohttp/client.py\", line 581, in _request | RuntimeError: Session is closed",
-      "n_samples": null,
-      "n_failed": null
-    },
-    "simpleqa": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T11:51:49.218418+00:00",
-      "raw": {
-        "correct": 0,
-        "incorrect": 0,
-        "not_attempted": 0,
-        "is_correct": 0.0,
-        "accuracy_given_attempted": 0.0,
-        "f1": 0.0
-      },
-      "judge_model": "claude-sonnet-4-6",
-      "error": "300/300 API calls failed after 6 retries (rate-limit/timeout/unsupported-param — see DLQ sample_errors); result unreliable",
-      "n_samples": 0,
-      "n_failed": 300
-    },
-    "strongreject": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T11:53:44.469755+00:00",
-      "raw": {
-        "asr": 0.0,
-        "n_scored": 0,
-        "evaluator": "strongreject_rubric"
-      },
-      "judge_model": "claude-sonnet-4-6",
-      "error": "313/313 API calls failed after 6 retries (rate-limit/timeout/unsupported-param — see DLQ sample_errors); result unreliable",
-      "n_samples": 0,
-      "n_failed": 313
-    },
-    "ethics": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T11:59:48.749639+00:00",
-      "raw": {},
-      "judge_model": null,
-      "error": "lm_eval exited 1 for tasks=ethics_justice_gen,ethics_deontology_gen,ethics_virtue_gen,ethics_utilitarianism_gen model=openai/gpt-5.5:     async with session.post( |                ^^^^^^^^^^^^^ |     ...<3 lines>... |     ) as response: |     ^ |   File \"/Users/vvettrivel/Documents/raidex-mono/backend/.venv/lib/python3.13/site-packages/aiohttp/client.py\", line 1683, in __aenter__ |   File \"/Users/vvettrivel/Documents/raidex-mono/backend/.venv/lib/python3.13/site-packages/aiohttp/client.py\", line 581, in _request | RuntimeError: Session is closed",
-      "n_samples": null,
-      "n_failed": null
-    },
-    "xstest": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T12:02:29.541393+00:00",
-      "raw": {
-        "safe_compliance": 0.0,
-        "unsafe_refusal": 0.0,
-        "safe_n": 0,
-        "unsafe_n": 0
-      },
-      "judge_model": "claude-sonnet-4-6",
-      "error": "450/450 API calls failed after 6 retries (rate-limit/timeout/unsupported-param — see DLQ sample_errors); result unreliable",
-      "n_samples": 0,
-      "n_failed": 450
-    },
-    "advglue": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T12:07:09.118570+00:00",
-      "raw": {},
-      "judge_model": null,
-      "error": "738/738 API calls failed after 6 retries (rate-limit/timeout/unsupported-param — see DLQ sample_errors); result unreliable",
-      "n_samples": 0,
-      "n_failed": 738
-    },
-    "confaide": {
-      "value": null,
-      "eval_source": "automated",
-      "eval_date": "2026-06-19T12:07:50.140394+00:00",
-      "raw": {
-        "pearson_tier_1": null,
-        "pearson_tier_2a": null,
-        "metric": "mean Pearson(model,human) mapped to (r+1)/2"
-      },
-      "judge_model": null,
-      "error": "108/108 API calls failed after 6 retries (rate-limit/timeout/unsupported-param — see DLQ sample_errors); result unreliable",
-      "n_samples": 0,
-      "n_failed": 108
-    }
-  },
-  "composite": {
-    "rai_score": null,
-    "rai_coverage": "0/8",
-    "rai_coverage_pct": 0,
-    "badge": "partial",
-    "badge_emoji": "⚪",
-    "dimension_scores": {
-      "safety": null,
-      "fairness_bias": null,
-      "factuality": null,
-      "security": null,
-      "robustness": null,
-      "privacy": null,
-      "machine_ethics": null
-    }
-  }
-}