ARCHIVED TRAINING / hgrecco/pint

hgrecco__pint-1111

This is a preserved training example. Membership in the current training wave has not been verified.

Problem statement

Context redefinitions can't be overlayed
When multiple contexts are active and more than one redefines the same unit, pint incorrectly picks up the first redefinition instead of the latest:

```python
ureg = pint.UnitRegistry()
c1 = pint.Context.from_lines(["@context a", "lb = 0.5 kg"])
c2 = pint.Context.from_lines(["@context b", "lb = 0.7 kg"])
one_src = ureg.Quantity(1, "kg")
print(one_src.to("lb"))
ureg.enable_contexts(c1)
print(one_src.to("lb"))
ureg.enable_contexts(c2)
print(one_src.to("lb"))
```
Expected output:
```
2.2046226218487757 pound
2.0 pound
1.4285714285714286 pound
```
Actual output:
```
2.2046226218487757 pound
2.0 pound
2.0 pound
```

Inter-dimensional transforms instead work as expected:
```python
ureg = pint.UnitRegistry()
c1 = pint.Context.from_lines(["@context a", "[volume] -> [mass]: value * 1 kg/l"])
c2 = pint.Context.from_lines(["@context b", "[volume] -> [mass]: value * 2 kg/l"])
one_src = ureg.Quantity(1, "l")
ureg.enable_contexts(c1)
print(one_src.to("kg"))
ureg.enable_contexts(c2)
print(one_src.to("kg"))
```
Output:
```
1.0 kilogram
2.0 kilogram
```

Training membership

opd_e4b_pure_v7 opd_e4b_pure_v7b

Task issues extracted from preserved verified_rft_messages training examples. This is not a verified current RL wave manifest.

Task and snapshot provenance

Snapshot captured 2026-11-12T17:06:59.476467+00:00 · Autoresearch-Arena/gemma-4-developer-agent

{
  "datasets": [
    "opd_e4b_pure_v7",
    "opd_e4b_pure_v7b"
  ],
  "examples": 9,
  "instance_id": "hgrecco__pint-1111",
  "problem_statement": "Context redefinitions can't be overlayed\nWhen multiple contexts are active and more than one redefines the same unit, pint incorrectly picks up the first redefinition instead of the latest:\r\n\r\n```python\r\nureg = pint.UnitRegistry()\r\nc1 = pint.Context.from_lines([\"@context a\", \"lb = 0.5 kg\"])\r\nc2 = pint.Context.from_lines([\"@context b\", \"lb = 0.7 kg\"])\r\none_src = ureg.Quantity(1, \"kg\")\r\nprint(one_src.to(\"lb\"))\r\nureg.enable_contexts(c1)\r\nprint(one_src.to(\"lb\"))\r\nureg.enable_contexts(c2)\r\nprint(one_src.to(\"lb\"))\r\n```\r\nExpected output:\r\n```\r\n2.2046226218487757 pound\r\n2.0 pound\r\n1.4285714285714286 pound\r\n```\r\nActual output:\r\n```\r\n2.2046226218487757 pound\r\n2.0 pound\r\n2.0 pound\r\n```\r\n\r\nInter-dimensional transforms instead work as expected:\r\n```python\r\nureg = pint.UnitRegistry()\r\nc1 = pint.Context.from_lines([\"@context a\", \"[volume] -> [mass]: value * 1 kg/l\"])\r\nc2 = pint.Context.from_lines([\"@context b\", \"[volume] -> [mass]: value * 2 kg/l\"])\r\none_src = ureg.Quantity(1, \"l\")\r\nureg.enable_contexts(c1)\r\nprint(one_src.to(\"kg\"))\r\nureg.enable_contexts(c2)\r\nprint(one_src.to(\"kg\"))\r\n```\r\nOutput:\r\n```\r\n1.0 kilogram\r\n2.0 kilogram\r\n```",
  "repo": "hgrecco/pint",
  "source": "v2/experiments/train_data/opd_e4b_pure_v7/verified_rft_messages.jsonl",
  "source_sha256": "e37f7bd675322f3f81500b73559d60d5f602ca0537bf19f5cb06332c728c7b72"
}

{
  "captured_at": "2026-11-12T17:06:59.476467+00:00",
  "membership": "archived training",
  "membership_evidence": "Task issues extracted from preserved verified_rft_messages training examples. This is not a verified current RL wave manifest.",
  "name": "Preserved post-training task archive",
  "source": "Autoresearch-Arena/gemma-4-developer-agent",
  "source_revision": "016496de222aaafca0d4ee4c18efc63ba616f76e",
  "sources": [
    {
      "path": "v2/experiments/train_data/opd_e4b_cispo_v5/verified_rft_messages.jsonl",
      "sha256": "a43610ad0dfb6ace95375cfae09000f41afc0338cc925af870a4ade7a9b20c4c"
    },
    {
      "path": "v2/experiments/train_data/opd_e4b_hybrid_v3/verified_rft_messages.jsonl",
      "sha256": "bd326303eb127e8e5e08f03daf97bb88d79d19709d59f3f93f348bc56686a8df"
    },
    {
      "path": "v2/experiments/train_data/opd_e4b_onpolicy_v3/verified_rft_messages.jsonl",
      "sha256": "960c680e797e823a0d3a22f4b1d5ef60a72971373d703b877340b0c9126aaf68"
    },
    {
      "path": "v2/experiments/train_data/opd_e4b_pure_v6a/verified_rft_messages.jsonl",
      "sha256": "f045b7ca5b3883a1e2eaee81e1c02a4a1b53dd7b7b36e98bacb2ba5823a15148"
    },
    {
      "path": "v2/experiments/train_data/opd_e4b_pure_v6b/verified_rft_messages.jsonl",
      "sha256": "0f62460a9706eea5e1605255896f191ab00cb5706c5f1f2e1c293e9a718218dc"
    },
    {
      "path": "v2/experiments/train_data/opd_e4b_pure_v7/verified_rft_messages.jsonl",
      "sha256": "e37f7bd675322f3f81500b73559d60d5f602ca0537bf19f5cb06332c728c7b72"
    },
    {
      "path": "v2/experiments/train_data/opd_e4b_pure_v7b/verified_rft_messages.jsonl",
      "sha256": "b864bdc8695183433eef6e31cb8d33fad44ba3aded2170c1ade4ad5b2944ce23"
    },
    {
      "path": "v2/experiments/train_data/opd_e4b_self_audit_v4/verified_rft_messages.jsonl",
      "sha256": "0026023745afb8f3dc1c176dc7f7171157b4ad5a66656b524af218e8e19ceee6"
    },
    {
      "path": "v2/experiments/train_data/rft_e4b_aligned_v2/verified_rft_messages.jsonl",
      "sha256": "92a4515fee9d152b0d4c510ee3ce4caaaffe2003ba7584469f8027f174aab18c"
    },
    {
      "path": "v2/experiments/train_data/rft_v1/verified_rft_messages.jsonl",
      "sha256": "a42453ccd7ac7680b0a03865e8e7080bc8bf74c8d9d193b440c441ed46f8512b"
    },
    {
      "path": "v2/experiments/train_data/rft_v2/verified_rft_messages.jsonl",
      "sha256": "18f4b3bd7e651f83c6acab479c9475e3fa50dc652a384f309e97146983ac85ba"
    }
  ]
}