Files
nl2jq-qwen3-0.6b/eval_results.json

26 lines
632 B
JSON
Raw Normal View History

{
"benchmark": "gauthierpiarrette/nl2jq-bench",
"version": "1.0.0",
"freeze_sha256": "25e7c01360707b30df93e7c1ed26997b3bb3ee7840a093d024bb4d16f489c5de",
"jq_version": "1.7.1",
"n_items": 400,
"protocol": "one-shot; no selection or iteration against the frozen set",
"model": "gauthierpiarrette/nl2jq-qwen3-0.6b",
"results": [
{
"system": "nl2jq-qwen3-0.6b-v6",
"decoding": "greedy",
"pass_at_1": 0.4,
"valid_at_1": 0.73,
"tiers": {
"T1": 0.65,
"T2": 0.52,
"T3": 0.36,
"T4": 0.28,
"T5": 0.2
},
"t5_generalization": 0.2
}
]
}