{"version":1,"run_id":"run_7511bacf6e","files":[{"name":"code-revision.txt","kind":"code_revision","sha256":"06bc234a4ce40000f1382dfbb653a243b7dfb3fa6e60d7739ac113262a4f9403","size":655,"url":"/api/runs/run_7511bacf6e/evidence/06bc234a4ce40000f1382dfbb653a243b7dfb3fa6e60d7739ac113262a4f9403"}],"checkpoints":[{"arm":"grpo","uri":"tinker://a14956b6cd9d6e01ce4bdf0b832d6fd8:train:0/sampler_weights/grpo-s14","step":14},{"arm":"sdpo_token","uri":"tinker://4bdcbef25e675e7b6db12b015d1a2dcd:train:0/sampler_weights/sdpo_token-s14","step":14}],"per_arm_metrics":{"base":{"composite_rubric_score":0.9583333333333334,"tau_bench_task_reward":0.05},"grpo":{"composite_rubric_score":0.96,"tau_bench_task_reward":0.05},"sdpo_token":{"composite_rubric_score":0.99,"tau_bench_task_reward":0.08333333333333333}},"missing":["script","effective_config","predictions"],"errors":["file 1: contains a literal credential assignment"],"status":"partial"}