{"version":1,"run_id":"run_bf5c66cfd2","files":[{"name":"launch-instructions-e43e9a9f1332.md","kind":"instructions","sha256":"e43e9a9f1332e586e71ab0a706e66398b66b7a7ccf61f96981b22092502a66f5","size":40507,"url":"/api/runs/run_bf5c66cfd2/evidence/e43e9a9f1332e586e71ab0a706e66398b66b7a7ccf61f96981b22092502a66f5"},{"name":"launch-instructions.md","kind":"instructions","sha256":"e43e9a9f1332e586e71ab0a706e66398b66b7a7ccf61f96981b22092502a66f5","size":40507,"url":"/api/runs/run_bf5c66cfd2/evidence/e43e9a9f1332e586e71ab0a706e66398b66b7a7ccf61f96981b22092502a66f5"},{"name":"full_run.py","kind":"script","sha256":"9819649af8246aae4ceb4b478800f37938fb0c933c32009d1d82e9b57a99acda","size":21665,"url":"/api/runs/run_bf5c66cfd2/evidence/9819649af8246aae4ceb4b478800f37938fb0c933c32009d1d82e9b57a99acda"},{"name":"rx_gs1.py","kind":"script","sha256":"fca7ac94282999a136228f2bd6fce7944a20c4ac21959bbb8774fb93aa824202","size":8096,"url":"/api/runs/run_bf5c66cfd2/evidence/fca7ac94282999a136228f2bd6fce7944a20c4ac21959bbb8774fb93aa824202"},{"name":"gs1_reference.py","kind":"script","sha256":"757d81c606f9793f259c6afd5ffe6f5e52382328560145b682d84d814da9a50e","size":1471,"url":"/api/runs/run_bf5c66cfd2/evidence/757d81c606f9793f259c6afd5ffe6f5e52382328560145b682d84d814da9a50e"},{"name":"eval.py","kind":"script","sha256":"bd1685b8ff81d0ebabbeffbdd645f470bf51f2d54b885018a738c6bfe81ca177","size":7520,"url":"/api/runs/run_bf5c66cfd2/evidence/bd1685b8ff81d0ebabbeffbdd645f470bf51f2d54b885018a738c6bfe81ca177"},{"name":"pilot.py","kind":"script","sha256":"3908eba8c54aca89ed9a5a7177cf7d8d7f51d0c431f1f1a72cc0a4f20e23e5b3","size":11258,"url":"/api/runs/run_bf5c66cfd2/evidence/3908eba8c54aca89ed9a5a7177cf7d8d7f51d0c431f1f1a72cc0a4f20e23e5b3"},{"name":"run_preflight.py","kind":"script","sha256":"fe8c51605bae70f58900597f10a2b2ee9c4b45aab55c2b8741fae4df38de0d1c","size":6674,"url":"/api/runs/run_bf5c66cfd2/evidence/fe8c51605bae70f58900597f10a2b2ee9c4b45aab55c2b8741fae4df38de0d1c"},{"name":"remote_stages.py","kind":"script","sha256":"2e00c243730f7020ec1b412a265325f6445fb62e4143a9ddbd4380a6f1039684","size":9463,"url":"/api/runs/run_bf5c66cfd2/evidence/2e00c243730f7020ec1b412a265325f6445fb62e4143a9ddbd4380a6f1039684"},{"name":"offline_tests.py","kind":"script","sha256":"c7ef1b9293f3f381c975215582546f717165b7831f0bf45d4098faa99bb6ed91","size":1634,"url":"/api/runs/run_bf5c66cfd2/evidence/c7ef1b9293f3f381c975215582546f717165b7831f0bf45d4098faa99bb6ed91"},{"name":"seed_probe.py","kind":"script","sha256":"4f79627f70bafc9bb3f5f7f36335d69fdd466177426ee0f8366359b99f811c55","size":829,"url":"/api/runs/run_bf5c66cfd2/evidence/4f79627f70bafc9bb3f5f7f36335d69fdd466177426ee0f8366359b99f811c55"},{"name":"assemble.py","kind":"script","sha256":"bd426601bd67f55b69e5e080563e7428667f5f4e3ce0402642586f40b425330e","size":5091,"url":"/api/runs/run_bf5c66cfd2/evidence/bd426601bd67f55b69e5e080563e7428667f5f4e3ce0402642586f40b425330e"},{"name":"effective-spec.json","kind":"effective_config","sha256":"97d70674ad162be54488b34ad460b11c8003f9819db5ee15bca9b82d099fa82a","size":12566,"url":"/api/runs/run_bf5c66cfd2/evidence/97d70674ad162be54488b34ad460b11c8003f9819db5ee15bca9b82d099fa82a"},{"name":"pins.json","kind":"code_revision","sha256":"63da79f31d56f4d783c4cbb0cd1c6f757b7cafe88b76e7e40e7c535aa6d68a07","size":642,"url":"/api/runs/run_bf5c66cfd2/evidence/63da79f31d56f4d783c4cbb0cd1c6f757b7cafe88b76e7e40e7c535aa6d68a07"},{"name":"decision-full.json","kind":"metrics","sha256":"d319e56a84e2be5446378feaf1dc7e802673e31c797828cdb97062f32db67620","size":595,"url":"/api/runs/run_bf5c66cfd2/evidence/d319e56a84e2be5446378feaf1dc7e802673e31c797828cdb97062f32db67620"},{"name":"paired-hoeffding.json","kind":"metrics","sha256":"fe10926501859e96291ffae4e4e9e44afa036e31b5c049932bd8a85d7abcc365","size":124,"url":"/api/runs/run_bf5c66cfd2/evidence/fe10926501859e96291ffae4e4e9e44afa036e31b5c049932bd8a85d7abcc365"},{"name":"preflight.json","kind":"metrics","sha256":"bbe4f0b5905664f1073cd15d9bedb67109e57b8ea17630f9f728aa23a78dda79","size":1860,"url":"/api/runs/run_bf5c66cfd2/evidence/bbe4f0b5905664f1073cd15d9bedb67109e57b8ea17630f9f728aa23a78dda79"},{"name":"validator-effective.json","kind":"metrics","sha256":"bfc1aa5ad2645455dc3a12444079fdb9bcaedd260971a3fcf47cf8e74dc63bad","size":1329,"url":"/api/runs/run_bf5c66cfd2/evidence/bfc1aa5ad2645455dc3a12444079fdb9bcaedd260971a3fcf47cf8e74dc63bad"},{"name":"validator-original.json","kind":"metrics","sha256":"3f219163eb7fdaeb84389e806b71dff7b5127a7e9673bc49a35140d556b82aaf","size":1371,"url":"/api/runs/run_bf5c66cfd2/evidence/3f219163eb7fdaeb84389e806b71dff7b5127a7e9673bc49a35140d556b82aaf"},{"name":"restoration-validation.json","kind":"metrics","sha256":"5251d7d723535739a967d3e9dca55d61778345bdf1dce76b56b59c0246370cb1","size":1605,"url":"/api/runs/run_bf5c66cfd2/evidence/5251d7d723535739a967d3e9dca55d61778345bdf1dce76b56b59c0246370cb1"},{"name":"initialization_provenance_full.json","kind":"metrics","sha256":"ef16e0790f16a457a3f5ae5093f1a2ecf0ce4bb989fe32342f1d51b4eeea61bd","size":4677,"url":"/api/runs/run_bf5c66cfd2/evidence/ef16e0790f16a457a3f5ae5093f1a2ecf0ce4bb989fe32342f1d51b4eeea61bd"},{"name":"initialization_provenance.json","kind":"metrics","sha256":"1b76fd7fca1d94b5b3f06530c397455093460d4da34fdc0bf501589478779649","size":5131,"url":"/api/runs/run_bf5c66cfd2/evidence/1b76fd7fca1d94b5b3f06530c397455093460d4da34fdc0bf501589478779649"},{"name":"probe.json","kind":"metrics","sha256":"57ef9aa15aaae87aa297ed04fb3d86848a42aa8f7ad6eb0226abe7ad4c1ef2a3","size":1277,"url":"/api/runs/run_bf5c66cfd2/evidence/57ef9aa15aaae87aa297ed04fb3d86848a42aa8f7ad6eb0226abe7ad4c1ef2a3"},{"name":"offline-advantage-tests.json","kind":"metrics","sha256":"a4ea95ead0bd7d20132ac482dd28b7ec5eb3295dce0f100c3e542bcc52e24d53","size":761,"url":"/api/runs/run_bf5c66cfd2/evidence/a4ea95ead0bd7d20132ac482dd28b7ec5eb3295dce0f100c3e542bcc52e24d53"},{"name":"activation-calibration.json","kind":"metrics","sha256":"dea7a52382ba5c6eb48b706da272d6fc27642d0ecc78a2c159be5e0b1d6d158d","size":1080,"url":"/api/runs/run_bf5c66cfd2/evidence/dea7a52382ba5c6eb48b706da272d6fc27642d0ecc78a2c159be5e0b1d6d158d"},{"name":"activation-pilot-gate-bw0.5.json","kind":"metrics","sha256":"672b1018ebae313c26c24b5de363cc609cd6e2f703972505d034a4f73b10f533","size":564,"url":"/api/runs/run_bf5c66cfd2/evidence/672b1018ebae313c26c24b5de363cc609cd6e2f703972505d034a4f73b10f533"},{"name":"activation-pilot-gate-bw1.0.json","kind":"metrics","sha256":"cd477aa621ec04f11d9b603d1139758a0c41b28c81b0e70fe9fc87e21d056aff","size":504,"url":"/api/runs/run_bf5c66cfd2/evidence/cd477aa621ec04f11d9b603d1139758a0c41b28c81b0e70fe9fc87e21d056aff"},{"name":"measurements.json","kind":"metrics","sha256":"93cf577efd57086d781efcd35cb3315582b74d2cdf77b41bdcc472da7d7bdbe3","size":1110,"url":"/api/runs/run_bf5c66cfd2/evidence/93cf577efd57086d781efcd35cb3315582b74d2cdf77b41bdcc472da7d7bdbe3"},{"name":"evaluation-base.json","kind":"metrics","sha256":"ce92325ada1c2b73351f5204206fd7cd70d251ca5911c929618c8f11f7965c7a","size":445,"url":"/api/runs/run_bf5c66cfd2/evidence/ce92325ada1c2b73351f5204206fd7cd70d251ca5911c929618c8f11f7965c7a"},{"name":"full__timings.json","kind":"metrics","sha256":"7cc19fd9f06d80643ed889b30619100e82387c1d6f0b29bace069750fa1980b5","size":121,"url":"/api/runs/run_bf5c66cfd2/evidence/7cc19fd9f06d80643ed889b30619100e82387c1d6f0b29bace069750fa1980b5"}],"checkpoints":[{"arm":"grpo_gs8","uri":"tinker://4a41c84719dfa5985f5ce122cf586538/weights/rx-full-grpo_gs8-final","step":40},{"arm":"grpo_gs8","uri":"tinker://4a41c84719dfa5985f5ce122cf586538:train:0/sampler_weights/rx-full-grpo_gs8-final","step":40},{"arm":"gs1_kernel","uri":"tinker://82134dc5c7344adfe7534612673e9353/weights/rx-full-gs1_kernel-final","step":40},{"arm":"gs1_kernel","uri":"tinker://82134dc5c7344adfe7534612673e9353:train:0/sampler_weights/rx-full-gs1_kernel-final","step":40},{"arm":"init","uri":"tinker://337d1e6ad746cc57412bd1332d4d628b/weights/rx-gs1-init-run_bf5c66cfd2","step":0}],"per_arm_metrics":{"grpo_gs8":{"accuracy":0.7134192570128886,"correct":941,"n_eval":1319,"steps":40,"mean_train_reward":0.6780078125,"frac_nonzero_advantage":0.696875,"mean_step_seconds":31.904820145925015,"executed_group_size":8.0},"gs1_kernel":{"accuracy":0.7460197119029568,"correct":984,"n_eval":1319,"steps":40,"mean_train_reward":0.67875,"frac_nonzero_advantage":1.0,"mean_step_seconds":25.02958305557502,"executed_group_size":1.0,"bandwidth":1.0,"min_step_kernel_neff_p10":20.546155946720656},"base":{"accuracy":0.7058377558756633,"correct":931,"n_eval":1319}},"missing":["predictions"],"errors":["file 24: invalid file name or kind","file 31: attachment retrieval returned HTTP 401","file 32: attachment retrieval returned HTTP 401","file 33: attachment retrieval returned HTTP 401","file 34: attachment retrieval returned HTTP 401","file 35: attachment retrieval returned HTTP 401","file 36: attachment retrieval returned HTTP 401","file 37: attachment retrieval returned HTTP 401","file 38: attachment retrieval returned HTTP 401","file 39: attachment retrieval returned HTTP 401","file 40: attachment retrieval returned HTTP 401"],"status":"partial"}