{ "regime": "pi1", "strategy": "grpo", "semantic_step": 500, "provenance": "Historical pi1 pure GRPO; matches per-checkpoint evaluation provenance.", "files_sha256": { "generation_config.json": "61c976d8443e6a7b4956b1847e4f970ab1ff5ad0db2b3ca3970b711c098c299f", "model-00002-of-00002.safetensors": "a6049b030aa0ea211a7090c5cfc70ee430f7fb027a378f0641fc31d8e5c6141e", "vocab.json": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910", "tokenizer_config.json": "6c90413caefc621f6fcfb8d9e5ce8ba9c949f5b65bcf8a4808f994fb9c51cf91", "merges.txt": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5", "special_tokens_map.json": "6676f091c8bc4d1b50146427cfde92073402866b87b6e39223227931b70083e9", "model.safetensors.index.json": "182b576f521e239d797460755941a1e4c8ffb838a83b0a6aa7dc449f942498df", "config.json": "f97ea6d1b921cd1e021be3d09a6d0bcc13a431d35bf6beb5333982d651011314", "tokenizer.json": "9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa", "model-00001-of-00002.safetensors": "42c9d4b5590281b10619866c4c88703b9f02b1bb38518324c3840b961365804b", "added_tokens.json": "58b54bbe36fc752f79a24a271ef66a0a0830054b4dfad94bde757d851968060b" }, "model_type": "qwen2" }