Upload mouse-run-run-2-mlp: trained agents, rollouts, evaluation, cross-architecture study

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+derived/reports/figures/evaluation_comparison.png filter=lfs diff=lfs merge=lfs -text
+derived/reports/figures/training_curves.png filter=lfs diff=lfs merge=lfs -text

HF_UPLOAD_MANIFEST.json

@@ -0,0 +1,799 @@
+{
+  "artifact_path": "runs/mouse-run-run-2-mlp",
+  "created_at": "2026-07-03T19:17:33.391331+00:00",
+  "excluded_from_file_manifest": [
+    "runs/mouse-run-run-2-mlp/HF_UPLOAD_MANIFEST.json"
+  ],
+  "file_count": 150,
+  "files": [
+    {
+      "path": "README.md",
+      "sha256": "0a0de8947639614dfb361bee8756fd16ccbd89d188cd749394e973a910793024",
+      "size_bytes": 2898
+    },
+    {
+      "path": "cross_architecture_study/cross_arch_cka.json",
+      "sha256": "1df526c3c10435203135df1ba02285eef813ebbaa176995178168313b6aeb797",
+      "size_bytes": 518
+    },
+    {
+      "path": "cross_architecture_study/cross_arch_perunit.json",
+      "sha256": "fda1247aa82af44396e5ab786e57622b4699290b6daf61319860442b72a4a852",
+      "size_bytes": 14851
+    },
+    {
+      "path": "cross_architecture_study/cross_arch_plsc_residualized.json",
+      "sha256": "596160749242ca86dee143e914c6c981654fa85bedb024f355c0a3236a909f70",
+      "size_bytes": 4062
+    },
+    {
+      "path": "cross_architecture_study/figures/cross_arch_cka.png",
+      "sha256": "7b43e1f229f0cec33101c1b89634082a5b84445e45e85ee114be19ad0c9c3dbb",
+      "size_bytes": 60221
+    },
+    {
+      "path": "cross_architecture_study/figures/sharing_vs_vision.png",
+      "sha256": "b9ddfbb26313e9013310fe5f7236dc50a5bc7ba4af2e8dff9c90331a64ad4f2a",
+      "size_bytes": 74823
+    },
+    {
+      "path": "cross_architecture_study/report.md",
+      "sha256": "26dde4d9e0348308c638aafa454f34404b3b81667fdfa24b24f67d47f015bf71",
+      "size_bytes": 11540
+    },
+    {
+      "path": "derived/reports/MANIFEST.json",
+      "sha256": "d838078800d08a887708e013f6ce920ef429c889b58dbf5d2ee6c718f5e6ee45",
+      "size_bytes": 2710
+    },
+    {
+      "path": "derived/reports/figures/evaluation_comparison.png",
+      "sha256": "f884d82cdc4bac267de1ab48120986784960a6448d8c7f4502cc3b48c4369735",
+      "size_bytes": 172530
+    },
+    {
+      "path": "derived/reports/figures/training_curves.png",
+      "sha256": "f79e51dd791199e40a8057983a80cb9405bc95f186a2e5ee61eae94544f41bac",
+      "size_bytes": 386863
+    },
+    {
+      "path": "derived/reports/report.md",
+      "sha256": "df7d44d955304bc3e6c45960376cc9cbb01a1a6fd836ad63804117a491620e8c",
+      "size_bytes": 3022
+    },
+    {
+      "path": "derived/reports/summary.json",
+      "sha256": "585e4bebe1633bf2d7b81c3bf38b86097c9586976bcbb28591044cbdf32f5152",
+      "size_bytes": 14882
+    },
+    {
+      "path": "derived/tables/MANIFEST.json",
+      "sha256": "bfe73384c93185ec5f56aff90884b5b6088e608289dc7efa88755a86d38b8edd",
+      "size_bytes": 2327
+    },
+    {
+      "path": "derived/tables/evaluations.jsonl",
+      "sha256": "265ed5550a98149da5384ac6cbcd551db5b121176f7046a4ca58d89531a77cc7",
+      "size_bytes": 135821
+    },
+    {
+      "path": "derived/tables/exclusions.jsonl",
+      "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+      "size_bytes": 0
+    },
+    {
+      "path": "derived/tables/panels.json",
+      "sha256": "cb73e84a077cbcd2d639c1c7f6947f3a418fa7482b62d1b789d5ad9608d30562",
+      "size_bytes": 760
+    },
+    {
+      "path": "derived/tables/rollouts.jsonl",
+      "sha256": "140147549967098f1a78dadab8e5888eb41a37389c094f79b623d22d7f0bd74a",
+      "size_bytes": 23011
+    },
+    {
+      "path": "derived/tables/runs.jsonl",
+      "sha256": "bfab762e5dd3673a8e08fa06635ebebd07bc02783f016fc26e56fee41756e2d8",
+      "size_bytes": 49200
+    },
+    {
+      "path": "eval.log",
+      "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+      "size_bytes": 0
+    },
+    {
+      "path": "launch_gate.json",
+      "sha256": "6e2fd3b528fbd3efe3328e8f44e7679fa619599ea2ea8af098f034831db3ca65",
+      "size_bytes": 4762
+    },
+    {
+      "path": "logs/non_social_seed_0000_attempt_01.log",
+      "sha256": "3a2b6a56180126c0db00b005208128dc688d7176074287a417a6048706b6235b",
+      "size_bytes": 9880
+    },
+    {
+      "path": "logs/non_social_seed_0001_attempt_01.log",
+      "sha256": "0aa37b6b0cd019a3d6309d0b91244b5b9b5371e29a40dabecebb253465aff7a2",
+      "size_bytes": 9864
+    },
+    {
+      "path": "logs/non_social_seed_0002_attempt_01.log",
+      "sha256": "ccc2bd907f4f26a2f4cf78e398e77a5ba40edf05d24fbbee4c3b7dd8083765de",
+      "size_bytes": 9882
+    },
+    {
+      "path": "logs/non_social_seed_0003_attempt_01.log",
+      "sha256": "d476bca205d0fb8c0ef9b4c1d141feb1fbdf10bde7549ce9c4f4abe04707f7e9",
+      "size_bytes": 9881
+    },
+    {
+      "path": "logs/non_social_seed_0004_attempt_01.log",
+      "sha256": "907770647fa99e4811ead15c437c5e168abda4ec9add43af615dfbb0d0f58b15",
+      "size_bytes": 9891
+    },
+    {
+      "path": "logs/non_social_seed_0005_attempt_01.log",
+      "sha256": "bf9cd3e1fd692ccf839ff21ef79ce5df0d509055b20bd875fd1b316ce53f0472",
+      "size_bytes": 9890
+    },
+    {
+      "path": "logs/non_social_seed_0006_attempt_01.log",
+      "sha256": "c246f36daaadde9b5b4d7b15dbd746d8709be8f3eecc5e571d457e004e0684bd",
+      "size_bytes": 9884
+    },
+    {
+      "path": "logs/non_social_seed_0007_attempt_01.log",
+      "sha256": "cd66e9a39b49a3fa81a06659bd3deaf83f1633d4e703bc8fee128ed917843e77",
+      "size_bytes": 9891
+    },
+    {
+      "path": "logs/non_social_seed_0008_attempt_01.log",
+      "sha256": "5dff30a9bf230401173ee2d01636d2aa93de7acabf658ad63293bfe7e6b41452",
+      "size_bytes": 9890
+    },
+    {
+      "path": "logs/non_social_seed_0009_attempt_01.log",
+      "sha256": "dea4f6c6a2089721690d705414c31c73109f620b10c6389f2f97752fd20ed2be",
+      "size_bytes": 9897
+    },
+    {
+      "path": "logs/social_seed_0000_attempt_01.log",
+      "sha256": "2986b8e85c9e56c87c016f21bc310c04d7d8116c19ba77f77db5bc3be2bb6e15",
+      "size_bytes": 9898
+    },
+    {
+      "path": "logs/social_seed_0001_attempt_01.log",
+      "sha256": "3455734d414ab8fbdc514ef9048e02b5bd395c48c06caa02fabce4c29fc68aa4",
+      "size_bytes": 9885
+    },
+    {
+      "path": "logs/social_seed_0002_attempt_01.log",
+      "sha256": "1554d4b94d8f373fbc07624ec01a5da600d7fe5ba0e20e3f79366658da1c8ac1",
+      "size_bytes": 9885
+    },
+    {
+      "path": "logs/social_seed_0003_attempt_01.log",
+      "sha256": "19c661ccbcfa51b889b754fe2889d7b16900a1c74e5a7bd921701005eb5e6c31",
+      "size_bytes": 9883
+    },
+    {
+      "path": "logs/social_seed_0004_attempt_01.log",
+      "sha256": "32b92be23199e2511a7bb487aed2ba1ae774858f4bdd3a00b262191e1da1706e",
+      "size_bytes": 9880
+    },
+    {
+      "path": "logs/social_seed_0005_attempt_01.log",
+      "sha256": "693f324e857090798faa9e31bfaa8f68f0ff52315a5019ba85dd225886a8e1ae",
+      "size_bytes": 9895
+    },
+    {
+      "path": "logs/social_seed_0006_attempt_01.log",
+      "sha256": "2a96ed1152e16dbf9cd9be397c102db33665cc3c8289d3ef3735e72af8e39186",
+      "size_bytes": 9900
+    },
+    {
+      "path": "logs/social_seed_0007_attempt_01.log",
+      "sha256": "846fdea329feb45345bd55b3059f4eaf5dd50c3e8a94296e14de8ce2f9efb6f0",
+      "size_bytes": 9898
+    },
+    {
+      "path": "logs/social_seed_0008_attempt_01.log",
+      "sha256": "92b4b1dacc715edec48941233f5b96dd2a3333b1850375122e4331c68285e170",
+      "size_bytes": 9872
+    },
+    {
+      "path": "logs/social_seed_0009_attempt_01.log",
+      "sha256": "a3ed2eae88f29997826bc60c90a32a6762cbd5cb32c214f5049a36257291d00f",
+      "size_bytes": 9860
+    },
+    {
+      "path": "manifest.json",
+      "sha256": "5bd2d31615b52c6a71b1e429aaa8700ed125ba9f9f1a530b8f33b8fe49925f8e",
+      "size_bytes": 14522
+    },
+    {
+      "path": "non_social/seed_0000/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "f8e5409f712499d04d8944b4ecd7070e6a1688c886ed4463cf04bf85fc9707a3",
+      "size_bytes": 11460376
+    },
+    {
+      "path": "non_social/seed_0000/attempt_01/config.json",
+      "sha256": "d0dd81d4e5d43bbad6fb2fb66c9fb53c3c58778d489ae11c0d278e3ac12abb98",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0000/attempt_01/metrics.jsonl",
+      "sha256": "f22bc9638550a95a739113e80fb0fab418aec865b6828d2c31e958688155a1b5",
+      "size_bytes": 54203
+    },
+    {
+      "path": "non_social/seed_0000/attempt_01/status.json",
+      "sha256": "671947ec08a3ba606c15d40fdfc2d648def59e8e0ade1ba19bd1647875276390",
+      "size_bytes": 1367
+    },
+    {
+      "path": "non_social/seed_0001/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "a6c8227e6a0fa824f91e1ee406af15f03ca2f65ee00d67445626986d7698c1c3",
+      "size_bytes": 11460368
+    },
+    {
+      "path": "non_social/seed_0001/attempt_01/config.json",
+      "sha256": "d80a42266c7a8dc06278f223dc4856658290ca6ca5de3b4d2db5e43084d98185",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0001/attempt_01/metrics.jsonl",
+      "sha256": "b3835354e5a75760d24115eef1ed938698d8f8c183687fea31af81c43b9a44d0",
+      "size_bytes": 54155
+    },
+    {
+      "path": "non_social/seed_0001/attempt_01/status.json",
+      "sha256": "f6dd4b15cd3e20fc0cd83bc84cec04dd038d36480b330f78a55caabb5f27caf1",
+      "size_bytes": 1358
+    },
+    {
+      "path": "non_social/seed_0002/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "e94060644c4de0c5467f9d15df5a5e4cfc6fbff18dcf358eb5cebe1dce94d3d7",
+      "size_bytes": 11460384
+    },
+    {
+      "path": "non_social/seed_0002/attempt_01/config.json",
+      "sha256": "f7bb8be082e2f25f89d008e4d0e3b9630dd24dce7af045a4018e95750b5f7a3e",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0002/attempt_01/metrics.jsonl",
+      "sha256": "a9283e7b6dbe7047f9ba222b9051cdeb19dbaa86198a68ca512710b21f94d7c5",
+      "size_bytes": 54093
+    },
+    {
+      "path": "non_social/seed_0002/attempt_01/status.json",
+      "sha256": "293c9cda446c7ea2f8c1cedff7793f052b70fe3abc31bd8c8758b026efb77539",
+      "size_bytes": 1371
+    },
+    {
+      "path": "non_social/seed_0003/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "17f68c1c23df5485beda4cdb0392ee32910c36897733023f84588c380da48191",
+      "size_bytes": 11460384
+    },
+    {
+      "path": "non_social/seed_0003/attempt_01/config.json",
+      "sha256": "0380ad17a8054b1c662e8cac4df015a96649e416dd5dc29eb0f46aff45917d5d",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0003/attempt_01/metrics.jsonl",
+      "sha256": "9ca6e1775ba7ef2367f12aadafe906e78c6a47a168f81bb08cb5f3f20678a1e9",
+      "size_bytes": 54134
+    },
+    {
+      "path": "non_social/seed_0003/attempt_01/status.json",
+      "sha256": "794957f5ae2c8c7b84da03deb9b15cd7b2492bff9fc701d850ffcce80c026b5a",
+      "size_bytes": 1374
+    },
+    {
+      "path": "non_social/seed_0004/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "1fa11ee31375fa76d13192833c9376692c520fcc925fee9c6b76a18abd4318fa",
+      "size_bytes": 11460392
+    },
+    {
+      "path": "non_social/seed_0004/attempt_01/config.json",
+      "sha256": "23629c5b243e4ce830a736a5e027e0f3d091a167e6c71bb0a58d4660dce00005",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0004/attempt_01/metrics.jsonl",
+      "sha256": "2ce1f9f59a65048a7b6f14af65e29d9ee9ac017471cf73a73f4bd24a159ca207",
+      "size_bytes": 54115
+    },
+    {
+      "path": "non_social/seed_0004/attempt_01/status.json",
+      "sha256": "d06cf0ff249fbe63f7a9a3cd31b5478ff83dca66ed511cf91310bc985aa4416c",
+      "size_bytes": 1380
+    },
+    {
+      "path": "non_social/seed_0005/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "5debd7cc50f687b861a1987d4433d6434b53e3ca6cd4c4743ac9192794e9ed0c",
+      "size_bytes": 11460392
+    },
+    {
+      "path": "non_social/seed_0005/attempt_01/config.json",
+      "sha256": "d14d2d282339aef9695bfdcfcd9323ba34a14012cba8d22f4c9bf8d3470c8986",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0005/attempt_01/metrics.jsonl",
+      "sha256": "23bb5e4f45b3bd07dde09c4e377826d37627fe941bcbd6a51d1c1a2086363efa",
+      "size_bytes": 54230
+    },
+    {
+      "path": "non_social/seed_0005/attempt_01/status.json",
+      "sha256": "40bd156060de2d8959b1160ed642e1430d337e6ab9243f748a2e0f5cb5766de8",
+      "size_bytes": 1383
+    },
+    {
+      "path": "non_social/seed_0006/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "01b8815629b4e75ed7baddd26bfade4a209f1c7c07e34b4bc26af47afcf2b7ed",
+      "size_bytes": 11460384
+    },
+    {
+      "path": "non_social/seed_0006/attempt_01/config.json",
+      "sha256": "65872e4b0c3720b0f541200245f545fc070a4e5c59ccad3488a4d2dc97510da1",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0006/attempt_01/metrics.jsonl",
+      "sha256": "b646d4a29dcb91f9901b4e4fa58b8b5b6e85e488d790251d69831f8449e3bdcb",
+      "size_bytes": 54179
+    },
+    {
+      "path": "non_social/seed_0006/attempt_01/status.json",
+      "sha256": "e483ebdb019fe6734731b01abf488a79ae8d49fb495e1695a5ba9a67fa7a511c",
+      "size_bytes": 1375
+    },
+    {
+      "path": "non_social/seed_0007/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "556bd8154b82d372e4f881e6319eeee0a1ad15f4e6c1b04df358d0711c48283b",
+      "size_bytes": 11460392
+    },
+    {
+      "path": "non_social/seed_0007/attempt_01/config.json",
+      "sha256": "41c29face687bbafdc78400210bb26a99b7c786300c9422ebcb4ea15eff03d32",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0007/attempt_01/metrics.jsonl",
+      "sha256": "2914bb836de2d92e9802974bc62aa62e848dd5d662de22743fde454e853270ed",
+      "size_bytes": 54109
+    },
+    {
+      "path": "non_social/seed_0007/attempt_01/status.json",
+      "sha256": "370e92fc6865eef1c1633b063798462e5a730f88b14ed595cd9ebdd193b1d894",
+      "size_bytes": 1384
+    },
+    {
+      "path": "non_social/seed_0008/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "d5ff79b8953d863166e3771c86afe125054881d88328fa27cd20f2770011ed51",
+      "size_bytes": 11460392
+    },
+    {
+      "path": "non_social/seed_0008/attempt_01/config.json",
+      "sha256": "9f4fd591b08685dbfb5dd5f6d91c4cdc9ba96a2b46e0163806a9135a7af8da0d",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0008/attempt_01/metrics.jsonl",
+      "sha256": "d80c0a1058016565bd18b54280b6cb7e5e87d399887079bcfa68813e1befca0d",
+      "size_bytes": 54226
+    },
+    {
+      "path": "non_social/seed_0008/attempt_01/status.json",
+      "sha256": "e79c1a7b7dc8db65c6d5b1b46713ccf52697d089b7a1f110355847c7e0f17c64",
+      "size_bytes": 1378
+    },
+    {
+      "path": "non_social/seed_0009/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "def8dbb315937927f72b78afdd9894dd5613ef88bc1bdbdb7b654ba5e17e690d",
+      "size_bytes": 11460400
+    },
+    {
+      "path": "non_social/seed_0009/attempt_01/config.json",
+      "sha256": "709f0eccca00218f98572be643caa8499ce9c3f2ba86cfeb3eedf3496f595209",
+      "size_bytes": 1185
+    },
+    {
+      "path": "non_social/seed_0009/attempt_01/metrics.jsonl",
+      "sha256": "b6d2c2e74baa3116060ab1bd83dc02fea726165ad25f4d91c1524a063211b281",
+      "size_bytes": 54154
+    },
+    {
+      "path": "non_social/seed_0009/attempt_01/status.json",
+      "sha256": "5932b45256cb8254b89d1be4fdd76e50a17d3b96cf11ccb2922223ae5a8d7559",
+      "size_bytes": 1386
+    },
+    {
+      "path": "paper_random_eval.jsonl",
+      "sha256": "287c7b3cbf320a6e79a47ff80c4bc086a7077076abe6ef0b800ebc663d2e6a3b",
+      "size_bytes": 132390
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0000__attempt_01__checkpoints__latest__9e74dbbabd.safetensors",
+      "sha256": "47235ca3adb8a5800cf9afc150039b2a465987c50ff85daa8b9f7e2eb91b4700",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0001__attempt_01__checkpoints__latest__6aac5a2c12.safetensors",
+      "sha256": "e6dcd1a1b640313bf009d7c3af192d2acfd254fa7e355e524c6f8ba35c4eae0a",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0002__attempt_01__checkpoints__latest__33f1a8c09e.safetensors",
+      "sha256": "6004b9a96caddd23e6c5370cb3a7dcbc79e7ccf86415bfc293f4f2859c725039",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0003__attempt_01__checkpoints__latest__ddb70257e3.safetensors",
+      "sha256": "a26129d1634332982cfc7436c50c3b54df91edd77569ff235c1a0d8f18507383",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0004__attempt_01__checkpoints__latest__8dc5f10a59.safetensors",
+      "sha256": "8001169fe126091e26b133edb59a13bb3bff90ab5826b6e32dd7745102385788",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0005__attempt_01__checkpoints__latest__c46a11793f.safetensors",
+      "sha256": "620dbd11a3b2f5485389a71091b2f45458b317db672910f107c567a98f03c28f",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0006__attempt_01__checkpoints__latest__c29253393a.safetensors",
+      "sha256": "cb1d4e549c50e6326007a139e6491233b34389321ae2fbab23d58825b7dbd8d4",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0007__attempt_01__checkpoints__latest__a64ce3bd0c.safetensors",
+      "sha256": "a941f04ea8dcce75abf0a42ed2329f1c987af0a7158fd5f92592eb67d418386e",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0008__attempt_01__checkpoints__latest__3f8cb5f7d2.safetensors",
+      "sha256": "3c249dac8de670dffeaf04b791c3b76f1f7015fad519831221eaf905d9e1b668",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0009__attempt_01__checkpoints__latest__e145c2cab5.safetensors",
+      "sha256": "ecd6db0f030df8ca8995578bb1416a8d00c4293749b545e0dbf7e583909903d4",
+      "size_bytes": 46506651
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0000__attempt_01__checkpoints__latest__f2865fcba9.safetensors",
+      "sha256": "0bad577c731d05df8b417b04bccfce215674bf0397bf0c87f6ea3f629e373f67",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0001__attempt_01__checkpoints__latest__61848b4a30.safetensors",
+      "sha256": "0e681e5199bb74bbb486c17420e6531bde8ca7f4873ba857ffd839b9e2e0bd1f",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0002__attempt_01__checkpoints__latest__77d0837868.safetensors",
+      "sha256": "2c1daa497f3a0b5503f0e5f7a87fea538c6dc3407fc2adf0dfa6d26bc726d5b4",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0003__attempt_01__checkpoints__latest__3d4e5109f0.safetensors",
+      "sha256": "ceb7f1689016b609d5e2713a0a2661ca9f83078a5aa89dd66851f17525c1f890",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0004__attempt_01__checkpoints__latest__4399b2519b.safetensors",
+      "sha256": "c12193945e93692204781680d7af3c920b1a0307a09e31072266eb2a5e4610db",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0005__attempt_01__checkpoints__latest__7ddcdb43f2.safetensors",
+      "sha256": "01465d07d01ed57f5c00a757c8a01ed8d23ccab455e35db5ae547eff48410011",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0006__attempt_01__checkpoints__latest__b87e617200.safetensors",
+      "sha256": "6433519f01ecece7e0023cf22dd156df34773a4064fd7e051275a0b34b41510f",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0007__attempt_01__checkpoints__latest__b494d062b3.safetensors",
+      "sha256": "86ef95b85618bcb31066bdd41f692e9b045da238f8e252197743daff8da6ac17",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0008__attempt_01__checkpoints__latest__c502c04b75.safetensors",
+      "sha256": "76c5e4773e0a78d321490080d05ddc09f3af942e00d6c8958cdb4e3d7cefcea4",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/mouse-run-run-2-mlp__social__seed_0009__attempt_01__checkpoints__latest__b92a4c3ad0.safetensors",
+      "sha256": "e98fced65d944fcdfc904a1b8344be4caaa933f30cb055861f623183b3df5dc9",
+      "size_bytes": 46506635
+    },
+    {
+      "path": "paper_rollouts/paper_rollout_records.jsonl",
+      "sha256": "bbe20472146213ba809436a6b0f28bfdcdad62b9e0f82bccd15d5e91346584c1",
+      "size_bytes": 19880
+    },
+    {
+      "path": "raw_records.jsonl",
+      "sha256": "be1e1692b0034a17180ddfc25d05bb02189b1cb558bd37060f90b8e283780445",
+      "size_bytes": 44596
+    },
+    {
+      "path": "report.log",
+      "sha256": "d838078800d08a887708e013f6ce920ef429c889b58dbf5d2ee6c718f5e6ee45",
+      "size_bytes": 2710
+    },
+    {
+      "path": "rollout.log",
+      "sha256": "0f815edad73fd2a4798725c22e7e2a8e2742646514fbc33277e9e46052dbac3f",
+      "size_bytes": 3060
+    },
+    {
+      "path": "run_status.json",
+      "sha256": "ef25ee077c79057a1fa041cbdb82bb7e7613546f5660088a82fb81ed3100f322",
+      "size_bytes": 38837
+    },
+    {
+      "path": "runner.log",
+      "sha256": "fbe8db5bee1737a5974e26e0fa79cdb31a4758bd9d61271010435804e07d6436",
+      "size_bytes": 3160
+    },
+    {
+      "path": "social/seed_0000/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "f4bcf1374937905b5d1f57c73129408305c6289a4aa0a054328102a1a4ceded0",
+      "size_bytes": 11460408
+    },
+    {
+      "path": "social/seed_0000/attempt_01/config.json",
+      "sha256": "26cc60a74e1f3e986a586451e1fbde856745fd6e48fba88062d1db3549d5fbd7",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0000/attempt_01/metrics.jsonl",
+      "sha256": "ced0dd7834936ea5020eda15921b74b58aafd9bfafde8e230083d97bc1055f23",
+      "size_bytes": 55340
+    },
+    {
+      "path": "social/seed_0000/attempt_01/status.json",
+      "sha256": "2e67a9fbca51f419e401e3cfff3ff4cc328bef207978a577e125df5f5e8f8c59",
+      "size_bytes": 1409
+    },
+    {
+      "path": "social/seed_0001/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "099cb185f62dd1f3f5f4d2842a1b0d283ade548522932f999c600e73c1ff44a8",
+      "size_bytes": 11460400
+    },
+    {
+      "path": "social/seed_0001/attempt_01/config.json",
+      "sha256": "184f4fc02e62aa29bc0d032edc7a322ae5c43832eab6a02947192b57c683b27a",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0001/attempt_01/metrics.jsonl",
+      "sha256": "ac0c73f137ab5242bcbb03dfdb981bb518828bfbaf2c1dffa44105d8db91021a",
+      "size_bytes": 55115
+    },
+    {
+      "path": "social/seed_0001/attempt_01/status.json",
+      "sha256": "cba00b2946080137b5be3675400144d40cdccbb1fab53fb153f5fd52770f1fb2",
+      "size_bytes": 1397
+    },
+    {
+      "path": "social/seed_0002/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "ddabd74e45c2c760a304a502299b61ff7db69c634e36f1f146feb9bf84e8fa5c",
+      "size_bytes": 11460400
+    },
+    {
+      "path": "social/seed_0002/attempt_01/config.json",
+      "sha256": "0b72006270f31460cf0317a5cc158b490fced4e44d7684a1ef9fdb560e4d749d",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0002/attempt_01/metrics.jsonl",
+      "sha256": "113a751b12158879e5b2b4f7500575c7f91f74edca88c16fcb2c5ab3cdcd7deb",
+      "size_bytes": 55101
+    },
+    {
+      "path": "social/seed_0002/attempt_01/status.json",
+      "sha256": "c1811a72c09743636b3bf2c3d657a00e0b827a5086f9dbe11d0b62abc82e1c92",
+      "size_bytes": 1398
+    },
+    {
+      "path": "social/seed_0003/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "77872aaef8ab25b68bf6d92a4af4bf74e5cc79b997d040bc34175db56af8e630",
+      "size_bytes": 11460400
+    },
+    {
+      "path": "social/seed_0003/attempt_01/config.json",
+      "sha256": "594facabfc3dad9237cbdb1bd1b0a4a87d7cb0cd6bd04d46ddf5f9179687f6de",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0003/attempt_01/metrics.jsonl",
+      "sha256": "048013e9095fe71db0cdf825b2d3d7acbb71e45af6a839527149e58023929d0e",
+      "size_bytes": 55169
+    },
+    {
+      "path": "social/seed_0003/attempt_01/status.json",
+      "sha256": "de6ff3ad58a9bbad450d69a2fef2a96ccb7f37277e0cb11356e313543ecd63ef",
+      "size_bytes": 1400
+    },
+    {
+      "path": "social/seed_0004/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "5f51e56c6790b8377549c656b1f50cce6f1bc65fa30c4ac612097554d1d0555f",
+      "size_bytes": 11460400
+    },
+    {
+      "path": "social/seed_0004/attempt_01/config.json",
+      "sha256": "66262f8aedc27441114a24a640b40c24428646f9b02e553a52e734df7210697f",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0004/attempt_01/metrics.jsonl",
+      "sha256": "5eb3af04e2ab4ab6bb5ac2c919c8a0cead0094136fd80aa8121aae6dbcda2b52",
+      "size_bytes": 55228
+    },
+    {
+      "path": "social/seed_0004/attempt_01/status.json",
+      "sha256": "7815b8148c28caa8e9b5bb63c088d64e7c2a87db14c9dab044f4d4a500926056",
+      "size_bytes": 1396
+    },
+    {
+      "path": "social/seed_0005/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "3405d3fc0a24dd17d23126ea788a9e1e3877444eb16623f74e0636865ace9e9b",
+      "size_bytes": 11460408
+    },
+    {
+      "path": "social/seed_0005/attempt_01/config.json",
+      "sha256": "416015f50c69ffef39d25e76049644951b1165dc8348c3b0353b8c78df1e65d2",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0005/attempt_01/metrics.jsonl",
+      "sha256": "20148b7f959cb2a43bc3e9eb513af5fe0339b16d32301243806b96092e6515b5",
+      "size_bytes": 55264
+    },
+    {
+      "path": "social/seed_0005/attempt_01/status.json",
+      "sha256": "23cf4d86baf4c7c4363e90431bfb5fd9f8733b1660a7154b27013a42db862f71",
+      "size_bytes": 1409
+    },
+    {
+      "path": "social/seed_0006/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "6a45c928ba31f02d9b235a4605d844df712be9103fbf93bfe039029aa9400fc4",
+      "size_bytes": 11460416
+    },
+    {
+      "path": "social/seed_0006/attempt_01/config.json",
+      "sha256": "7c5d18581b89ca5df4c1d4be8ebe8e9ba37fccc43ea3726c3dc8c97b795ceaab",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0006/attempt_01/metrics.jsonl",
+      "sha256": "516cf1adbc9c0075b8d9ebf320cd055f3d4b1d87db5376b1688db84120d57840",
+      "size_bytes": 55234
+    },
+    {
+      "path": "social/seed_0006/attempt_01/status.json",
+      "sha256": "97c600750a655b3c59a2eb1148d26f38e4ae8dfce850f91a71bf74b0726348fe",
+      "size_bytes": 1413
+    },
+    {
+      "path": "social/seed_0007/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "cb8a8e1bcbbb446b3ce500375218bbb11f70265bebe4d228299de67891943d15",
+      "size_bytes": 11460408
+    },
+    {
+      "path": "social/seed_0007/attempt_01/config.json",
+      "sha256": "1c3afe0ad677f8ae4c0ba672e2a4e76cd4b6c4f8857c0784cafb10b08a911944",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0007/attempt_01/metrics.jsonl",
+      "sha256": "4b67cc780716c9ef5557548c3396fed808c8b874ba5251b49412ee395ca8d39d",
+      "size_bytes": 55253
+    },
+    {
+      "path": "social/seed_0007/attempt_01/status.json",
+      "sha256": "f1676b961b553176b60650a6f26c591343b93c0f2b903d8eb311aea11dded755",
+      "size_bytes": 1408
+    },
+    {
+      "path": "social/seed_0008/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "39e2f67475bf8a4eca5f24de315185f9f759d1f65ac80f68a5644a58f16272e1",
+      "size_bytes": 11460392
+    },
+    {
+      "path": "social/seed_0008/attempt_01/config.json",
+      "sha256": "556e53904cd5c3e266faec185887090adf5d21612f4be7ad318482c0aafb871d",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0008/attempt_01/metrics.jsonl",
+      "sha256": "4e66fffd1375d22dbb32b2ec23e46b4b9519a9bf1cbcdf27e9104b740eaac813",
+      "size_bytes": 55240
+    },
+    {
+      "path": "social/seed_0008/attempt_01/status.json",
+      "sha256": "850d49a6859a22f969c245bdbea4c3a155a8b02fc6a3a7a4e90e49316130e5e9",
+      "size_bytes": 1387
+    },
+    {
+      "path": "social/seed_0009/attempt_01/checkpoints/latest.safetensors",
+      "sha256": "6790e80fe441a2995b3c9817361ed5add9b0fbc14ca34187b08685868aad6875",
+      "size_bytes": 11460376
+    },
+    {
+      "path": "social/seed_0009/attempt_01/config.json",
+      "sha256": "71afc05bd0004920ad0b97efb94bc3792c6ace32a2f13ccd3e1f8a39cba16d7d",
+      "size_bytes": 1172
+    },
+    {
+      "path": "social/seed_0009/attempt_01/metrics.jsonl",
+      "sha256": "08b517e6e73bc0d339c0166dbca4d8ee03e1ffd4b176c5bd30278ee2b9bb930e",
+      "size_bytes": 55147
+    },
+    {
+      "path": "social/seed_0009/attempt_01/status.json",
+      "sha256": "8039fd890ceb6927a99458c8ec08f1f466eef28e2dd18ea2f30d861da9043807",
+      "size_bytes": 1373
+    },
+    {
+      "path": "tables.log",
+      "sha256": "bfe73384c93185ec5f56aff90884b5b6088e608289dc7efa88755a86d38b8edd",
+      "size_bytes": 2327
+    },
+    {
+      "path": "triton_equivalence.json",
+      "sha256": "6fa7c8f5a90df05c0af5f7759f369fea3a0e468f998353df8f6f8b1b91a2ad99",
+      "size_bytes": 1935
+    }
+  ],
+  "path_in_repo": "",
+  "private": false,
+  "provenance": {
+    "command": "/Users/jacoblincool/Documents/GitHub/mouse-run-run/.venv/bin/upload-marl-artifacts runs/mouse-run-run-2-mlp --repo-id JacobLinCool/mouse-run-run-2-mlp --repo-type model --commit-message 'Upload mouse-run-run-2-mlp: trained agents, rollouts, evaluation, cross-architecture study'",
+    "created_at": "2026-07-03T19:17:33.857630+00:00",
+    "dependencies": {
+      "huggingface_hub": "1.21.0",
+      "numpy": "2.5.0",
+      "safetensors": "0.8.0",
+      "tensorboard": "2.21.0",
+      "torch": "2.12.1",
+      "triton": null
+    },
+    "executable": "/Users/jacoblincool/Documents/GitHub/mouse-run-run/.venv/bin/python3",
+    "git": {
+      "branch": "main",
+      "diff_sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+      "dirty": false,
+      "sha": "043425050d6ebb6201f0e8be035085405cb044ea",
+      "short_sha": "0434250",
+      "status_short": ""
+    },
+    "platform": "macOS-26.4-arm64-arm-64bit-Mach-O",
+    "python": "3.14.4 (main, Apr 14 2026, 14:46:33) [Clang 22.1.3 ]",
+    "schema_version": 1,
+    "torch": {
+      "cuda_available": false,
+      "cuda_device_count": 0,
+      "cuda_devices": [],
+      "cuda_version": null,
+      "cudnn_version": null,
+      "mps_available": true,
+      "version": "2.12.1"
+    }
+  },
+  "repo_id": "JacobLinCool/mouse-run-run-2-mlp",
+  "repo_type": "model",
+  "revision": null,
+  "schema_version": 1
+}

README.md

@@ -0,0 +1,71 @@
+---
+license: mit
+tags:
+- multi-agent-reinforcement-learning
+- ppo
+- recurrent-neural-network
+- neuroscience
+- inter-brain-dynamics
+---
+
+# mouse-run-run — MLP agents (mouse-run-run-2-mlp)
+
+Trained chaser/explorer agents for a modern PyTorch reproduction and
+architecture extension of the MARL experiment in Zhang et al. (2025),
+*Inter-brain neural dynamics in biological and artificial intelligence
+systems*, Nature 645, 991–1001.
+
+**Architecture:** frame-stack MLP (k=8, memoryless — common-input floor control). Analysis (hidden) dimension 256; two independently
+parameterized actor-critic agents (chaser and explorer, no weight sharing).
+
+This repo holds one full experiment: **20 trained agent pairs** (tasks
+`social` and `non_social` × seeds 0–9), each trained with PPO for 20,000
+updates × 40 episodes × 100 steps. 20/20 units completed on the first attempt.
+
+## Contents
+
+- `{social,non_social}/seed_XXXX/attempt_01/checkpoints/latest.safetensors` —
+  trained chaser+explorer weights (safetensors; config/metrics in metadata).
+- `paper_rollouts/*.safetensors` — 25×500-step analysis rollouts per pair:
+  per-timestep hidden states, positions, actions, and social-event flags
+  (schema v3, time-aligned; degenerate episodes flagged).
+- `paper_random_eval.jsonl` — standardized random-opponent evaluation.
+- `derived/{tables,reports}/` — canonical tables and the per-experiment report.
+- `cross_architecture_study/` — the four-architecture comparison report,
+  figures, and cross-architecture PLSC/CKA data.
+- `manifest.json`, `raw_records.jsonl`, `launch_gate.json`,
+  `triton_equivalence.json` — provenance.
+
+## Key results (this architecture)
+
+Behavior (vs standardized random opponent, social task):
+- chaser collisions / episode: **26.93**
+  (non-social control: 2.68)
+- chaser partner-in-vision: 0.93
+
+Neural (social agents):
+- collision decoding (balanced acc): **0.91**
+- partner escape / approach decoding: 0.78 / 0.78
+- PLSC shared-dimension top correlation: 0.68
+
+See `cross_architecture_study/report.md` for the four-architecture comparison.
+**Headline:** the architectures do not learn the same internal representations —
+only the RNN develops genuine internal shared dynamics at low mutual vision.
+
+## Load a checkpoint
+
+```python
+from huggingface_hub import hf_hub_download
+from mouse_run_run.serialization import load_checkpoint
+from mouse_run_run.policy import build_policy
+
+path = hf_hub_download("JacobLinCool/mouse-run-run-2-mlp",
+    "social/seed_0000/attempt_01/checkpoints/latest.safetensors")
+config, metrics, chaser_state, explorer_state = load_checkpoint(path)
+chaser = build_policy(config["architecture"], config["env"]["observation_size"],
+                      hidden_size=config["hidden_size"])
+chaser.load_state_dict(chaser_state)
+```
+
+Code: https://github.com/JacobLinCool/mouse-run-run (paper reproduction +
+cross-architecture study).

cross_architecture_study/cross_arch_cka.json

@@ -0,0 +1,18 @@
+{
+ "RNN|RNN": 1.0,
+ "RNN|MLP": 0.6792286610675173,
+ "RNN|SSM": 0.6359547814617931,
+ "RNN|Transformer": 0.6660120097138259,
+ "MLP|RNN": 0.6792286610675173,
+ "MLP|MLP": 1.0,
+ "MLP|SSM": 0.7110754305522333,
+ "MLP|Transformer": 0.7809858975192786,
+ "SSM|RNN": 0.6359547814617931,
+ "SSM|MLP": 0.7110754305522333,
+ "SSM|SSM": 1.0,
+ "SSM|Transformer": 0.7551091733153037,
+ "Transformer|RNN": 0.6660120097138259,
+ "Transformer|MLP": 0.7809858975192786,
+ "Transformer|SSM": 0.7551091733153037,
+ "Transformer|Transformer": 1.0
+}

cross_architecture_study/cross_arch_perunit.json

@@ -0,0 +1,802 @@
+[
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 7.440000057220459,
+  "top_r": 0.03705123940656156
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 10.039999961853027,
+  "top_r": 0.09726206921336848
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 12.199999809265137,
+  "top_r": 0.09102936182631945
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 3,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 19.8799991607666,
+  "top_r": 0.1416026760369321
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 9.880000114440918,
+  "top_r": 0.06074593045795309
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 5,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 11.760000228881836,
+  "top_r": 0.06100869691148739
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 11.760000228881836,
+  "top_r": 0.047992243964107055
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 7,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 9.720000267028809,
+  "top_r": 0.040861022303547455
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 8,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 8.359999656677246,
+  "top_r": 0.046082531832419726
+ },
+ {
+  "arch": "RNN",
+  "task": "non_social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 15.119999885559082,
+  "top_r": 0.04697112454078715
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 16,
+  "vision": 0.16239520907402039,
+  "collisions": 60.91999816894531,
+  "top_r": 0.9606705270007906
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 18,
+  "vision": 0.21668662130832672,
+  "collisions": 0.800000011920929,
+  "top_r": 0.6658904391428773
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 22,
+  "vision": 0.3185628652572632,
+  "collisions": 5.519999980926514,
+  "top_r": 0.6149421028031533
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 3,
+  "status": "degenerate",
+  "n_valid": 1,
+  "vision": 0.24694611132144928,
+  "collisions": 47.279998779296875,
+  "top_r": null
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 23,
+  "vision": 0.9924950003623962,
+  "collisions": 150.9199981689453,
+  "top_r": 0.9469448413117497
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 5,
+  "status": "ok",
+  "n_valid": 23,
+  "vision": 0.03936127573251724,
+  "collisions": 0.5199999809265137,
+  "top_r": 0.24632233096607034
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.2863073945045471,
+  "collisions": 1.1200000047683716,
+  "top_r": 0.6195669473643807
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 7,
+  "status": "degenerate",
+  "n_valid": 2,
+  "vision": 0.2092614769935608,
+  "collisions": 21.84000015258789,
+  "top_r": null
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 8,
+  "status": "ok",
+  "n_valid": 14,
+  "vision": 0.13309381902217865,
+  "collisions": 19.959999084472656,
+  "top_r": 0.9244864792894123
+ },
+ {
+  "arch": "RNN",
+  "task": "social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.2818363308906555,
+  "collisions": 5.960000038146973,
+  "top_r": 0.5286762777215525
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 8.279999732971191,
+  "top_r": 0.07776644999751171
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 8.640000343322754,
+  "top_r": 0.07086241042694791
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 9.039999961853027,
+  "top_r": 0.10571891551983957
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 3,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 10.600000381469727,
+  "top_r": 0.12438404297121783
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 13.15999984741211,
+  "top_r": 0.08272535053372708
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 5,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 4.599999904632568,
+  "top_r": 0.05911174780800376
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 6.599999904632568,
+  "top_r": 0.12048705520440492
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 7,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 15.720000267028809,
+  "top_r": 0.20280221422494873
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 8,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 13.640000343322754,
+  "top_r": 0.1197033617321138
+ },
+ {
+  "arch": "MLP",
+  "task": "non_social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 11.920000076293945,
+  "top_r": 0.06501230641904866
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.981237530708313,
+  "collisions": 115.08000183105469,
+  "top_r": 0.68309682518684
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9638323187828064,
+  "collisions": 114.72000122070312,
+  "top_r": 0.6770425529610226
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9689421057701111,
+  "collisions": 105.16000366210938,
+  "top_r": 0.6294873925115205
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 3,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9881038069725037,
+  "collisions": 236.0800018310547,
+  "top_r": 0.7859041502514537
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9528942108154297,
+  "collisions": 125.0,
+  "top_r": 0.7282436596506869
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 5,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9758083820343018,
+  "collisions": 110.36000061035156,
+  "top_r": 0.6151424539297031
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9893013834953308,
+  "collisions": 114.23999786376953,
+  "top_r": 0.7531067455891379
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 7,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9883432984352112,
+  "collisions": 141.52000427246094,
+  "top_r": 0.6580303971938998
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 8,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9682235717773438,
+  "collisions": 101.19999694824219,
+  "top_r": 0.6171618473537643
+ },
+ {
+  "arch": "MLP",
+  "task": "social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.9904990196228027,
+  "collisions": 115.23999786376953,
+  "top_r": 0.639261743430404
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.0,
+  "collisions": 40.2400016784668,
+  "top_r": 0.28866651056915354
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 3.4800000190734863,
+  "top_r": 0.15885310687782797
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 3.799999952316284,
+  "top_r": 0.1087223462156361
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 3,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 1.8799999952316284,
+  "top_r": 0.0647609791042199
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 34.36000061035156,
+  "top_r": 0.6156083569490233
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 5,
+  "status": "degenerate",
+  "n_valid": 1,
+  "vision": 0.0,
+  "collisions": 34.68000030517578,
+  "top_r": null
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.0,
+  "collisions": 3.200000047683716,
+  "top_r": 0.08990168984095623
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 7,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 121.23999786376953,
+  "top_r": 0.6383316081199204
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 8,
+  "status": "degenerate",
+  "n_valid": 0,
+  "vision": 0.0,
+  "collisions": 46.31999969482422,
+  "top_r": null
+ },
+ {
+  "arch": "SSM",
+  "task": "non_social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 105.68000030517578,
+  "top_r": 0.5532722247875926
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 13,
+  "vision": 0.9386028051376343,
+  "collisions": 107.16000366210938,
+  "top_r": 0.6762547587766922
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 19,
+  "vision": 0.6916567087173462,
+  "collisions": 41.91999816894531,
+  "top_r": 0.5728661391129467
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 22,
+  "vision": 0.9957684874534607,
+  "collisions": 63.91999816894531,
+  "top_r": 0.5386769468869881
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 3,
+  "status": "degenerate",
+  "n_valid": 0,
+  "vision": 0.9954491257667542,
+  "collisions": 400.9200134277344,
+  "top_r": null
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 5,
+  "vision": 0.4332934021949768,
+  "collisions": 20.15999984741211,
+  "top_r": 0.6451686137949718
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 5,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.981157660484314,
+  "collisions": 66.23999786376953,
+  "top_r": 0.5719388049785209
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.9827544689178467,
+  "collisions": 63.79999923706055,
+  "top_r": 0.6399909552314786
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 7,
+  "status": "ok",
+  "n_valid": 8,
+  "vision": 0.9846706390380859,
+  "collisions": 129.55999755859375,
+  "top_r": 0.5951176365769322
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 8,
+  "status": "ok",
+  "n_valid": 3,
+  "vision": 0.8347305655479431,
+  "collisions": 139.27999877929688,
+  "top_r": 0.5911175381223174
+ },
+ {
+  "arch": "SSM",
+  "task": "social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 8,
+  "vision": 0.9699001908302307,
+  "collisions": 279.9599914550781,
+  "top_r": 0.9463852317330589
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.0,
+  "collisions": 18.15999984741211,
+  "top_r": 0.20624045007674752
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 177.52000427246094,
+  "top_r": 0.5466457412672473
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 247.67999267578125,
+  "top_r": 0.741521978186391
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 3,
+  "status": "ok",
+  "n_valid": 12,
+  "vision": 0.0,
+  "collisions": 32.47999954223633,
+  "top_r": 0.29652097824664553
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 89.83999633789062,
+  "top_r": 0.3499635714443735
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 5,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 118.91999816894531,
+  "top_r": 0.4220079232087302
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 15,
+  "vision": 0.0,
+  "collisions": 199.83999633789062,
+  "top_r": 0.4868188359531619
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 7,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 87.63999938964844,
+  "top_r": 0.3143182076125383
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 8,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 12.920000076293945,
+  "top_r": 0.10559527101392678
+ },
+ {
+  "arch": "Transformer",
+  "task": "non_social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.0,
+  "collisions": 4.920000076293945,
+  "top_r": 0.0921079793072239
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 0,
+  "status": "ok",
+  "n_valid": 23,
+  "vision": 0.8109381198883057,
+  "collisions": 66.27999877929688,
+  "top_r": 0.4658015424447193
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 1,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.816926121711731,
+  "collisions": 83.04000091552734,
+  "top_r": 0.37708898276609976
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 2,
+  "status": "ok",
+  "n_valid": 14,
+  "vision": 0.9682235717773438,
+  "collisions": 81.19999694824219,
+  "top_r": 0.4850167809908031
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 3,
+  "status": "ok",
+  "n_valid": 20,
+  "vision": 0.8345708847045898,
+  "collisions": 78.55999755859375,
+  "top_r": 0.4330689299750323
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 4,
+  "status": "ok",
+  "n_valid": 21,
+  "vision": 0.757125735282898,
+  "collisions": 76.16000366210938,
+  "top_r": 0.34081183250426467
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 5,
+  "status": "ok",
+  "n_valid": 20,
+  "vision": 0.6459081768989563,
+  "collisions": 100.72000122070312,
+  "top_r": 0.4682438954043613
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 6,
+  "status": "ok",
+  "n_valid": 25,
+  "vision": 0.7647106051445007,
+  "collisions": 47.84000015258789,
+  "top_r": 0.37878757497109683
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 7,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.9699800610542297,
+  "collisions": 95.5999984741211,
+  "top_r": 0.4278219769902787
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 8,
+  "status": "ok",
+  "n_valid": 20,
+  "vision": 0.7192016243934631,
+  "collisions": 103.23999786376953,
+  "top_r": 0.6244838845224036
+ },
+ {
+  "arch": "Transformer",
+  "task": "social",
+  "seed": 9,
+  "status": "ok",
+  "n_valid": 24,
+  "vision": 0.7394012212753296,
+  "collisions": 95.95999908447266,
+  "top_r": 0.3750948684332812
+ }
+]

cross_architecture_study/cross_arch_plsc_residualized.json

@@ -0,0 +1,401 @@
+{
+ "RNN": {
+  "social": [
+   [
+    0.9606705270007906,
+    156,
+    16
+   ],
+   [
+    0.6658904391428773,
+    236,
+    18
+   ],
+   [
+    0.6149421028031533,
+    250,
+    22
+   ],
+   [
+    0.9469448413117497,
+    256,
+    23
+   ],
+   [
+    0.24632233096607034,
+    190,
+    23
+   ],
+   [
+    0.6195669473643807,
+    255,
+    25
+   ],
+   [
+    0.9244864792894123,
+    212,
+    14
+   ],
+   [
+    0.5286762777215525,
+    256,
+    24
+   ]
+  ],
+  "non_social": [
+   [
+    0.03705123940656156,
+    54,
+    25
+   ],
+   [
+    0.09726206921336848,
+    111,
+    25
+   ],
+   [
+    0.09102936182631945,
+    108,
+    25
+   ],
+   [
+    0.1416026760369321,
+    116,
+    25
+   ],
+   [
+    0.06074593045795309,
+    95,
+    25
+   ],
+   [
+    0.06100869691148739,
+    122,
+    25
+   ],
+   [
+    0.047992243964107055,
+    101,
+    25
+   ],
+   [
+    0.040861022303547455,
+    113,
+    25
+   ],
+   [
+    0.046082531832419726,
+    120,
+    25
+   ],
+   [
+    0.04697112454078715,
+    121,
+    25
+   ]
+  ]
+ },
+ "MLP": {
+  "social": [
+   [
+    0.68309682518684,
+    256,
+    25
+   ],
+   [
+    0.6770425529610226,
+    256,
+    25
+   ],
+   [
+    0.6294873925115205,
+    256,
+    25
+   ],
+   [
+    0.7859041502514537,
+    256,
+    25
+   ],
+   [
+    0.7282436596506869,
+    256,
+    25
+   ],
+   [
+    0.6151424539297031,
+    256,
+    25
+   ],
+   [
+    0.7531067455891379,
+    256,
+    25
+   ],
+   [
+    0.6580303971938998,
+    256,
+    25
+   ],
+   [
+    0.6171618473537643,
+    256,
+    25
+   ],
+   [
+    0.639261743430404,
+    256,
+    25
+   ]
+  ],
+  "non_social": [
+   [
+    0.07776644999751171,
+    253,
+    25
+   ],
+   [
+    0.07086241042694791,
+    254,
+    25
+   ],
+   [
+    0.10571891551983957,
+    255,
+    25
+   ],
+   [
+    0.12438404297121783,
+    255,
+    25
+   ],
+   [
+    0.08272535053372708,
+    255,
+    25
+   ],
+   [
+    0.05911174780800376,
+    255,
+    25
+   ],
+   [
+    0.12048705520440492,
+    251,
+    25
+   ],
+   [
+    0.20280221422494873,
+    256,
+    25
+   ],
+   [
+    0.1197033617321138,
+    231,
+    25
+   ],
+   [
+    0.06501230641904866,
+    251,
+    25
+   ]
+  ]
+ },
+ "SSM": {
+  "social": [
+   [
+    0.6762547587766922,
+    253,
+    13
+   ],
+   [
+    0.5728661391129467,
+    256,
+    19
+   ],
+   [
+    0.5386769468869881,
+    249,
+    22
+   ],
+   [
+    0.6451686137949718,
+    249,
+    5
+   ],
+   [
+    0.5719388049785209,
+    256,
+    25
+   ],
+   [
+    0.6399909552314786,
+    256,
+    24
+   ],
+   [
+    0.5951176365769322,
+    239,
+    8
+   ],
+   [
+    0.5911175381223174,
+    179,
+    3
+   ],
+   [
+    0.9463852317330589,
+    224,
+    8
+   ]
+  ],
+  "non_social": [
+   [
+    0.28866651056915354,
+    235,
+    24
+   ],
+   [
+    0.15885310687782797,
+    137,
+    25
+   ],
+   [
+    0.1087223462156361,
+    99,
+    25
+   ],
+   [
+    0.0647609791042199,
+    185,
+    25
+   ],
+   [
+    0.6156083569490233,
+    244,
+    25
+   ],
+   [
+    0.08990168984095623,
+    100,
+    24
+   ],
+   [
+    0.6383316081199204,
+    252,
+    25
+   ],
+   [
+    0.5532722247875926,
+    235,
+    25
+   ]
+  ]
+ },
+ "Transformer": {
+  "social": [
+   [
+    0.4658015424447193,
+    251,
+    23
+   ],
+   [
+    0.37708898276609976,
+    251,
+    24
+   ],
+   [
+    0.4850167809908031,
+    253,
+    14
+   ],
+   [
+    0.4330689299750323,
+    254,
+    20
+   ],
+   [
+    0.34081183250426467,
+    249,
+    21
+   ],
+   [
+    0.4682438954043613,
+    241,
+    20
+   ],
+   [
+    0.37878757497109683,
+    250,
+    25
+   ],
+   [
+    0.4278219769902787,
+    249,
+    24
+   ],
+   [
+    0.6244838845224036,
+    249,
+    20
+   ],
+   [
+    0.3750948684332812,
+    248,
+    24
+   ]
+  ],
+  "non_social": [
+   [
+    0.20624045007674752,
+    20,
+    24
+   ],
+   [
+    0.5466457412672473,
+    152,
+    25
+   ],
+   [
+    0.741521978186391,
+    139,
+    25
+   ],
+   [
+    0.29652097824664553,
+    16,
+    12
+   ],
+   [
+    0.3499635714443735,
+    91,
+    25
+   ],
+   [
+    0.4220079232087302,
+    70,
+    25
+   ],
+   [
+    0.4868188359531619,
+    209,
+    15
+   ],
+   [
+    0.3143182076125383,
+    181,
+    25
+   ],
+   [
+    0.10559527101392678,
+    113,
+    25
+   ],
+   [
+    0.0921079793072239,
+    15,
+    25
+   ]
+  ]
+ }
+}

cross_architecture_study/report.md

@@ -0,0 +1,215 @@
+# Cross-Architecture Representation Study
+
+**Question.** Do different agent architectures, trained identically on the
+Zhang et al. (2025) chaser–explorer task, learn the *same* internal
+representations of the social interaction? Or does the phenomenon the paper
+reports — genuine shared neural dynamics between two independently trained
+agents — depend on the architecture?
+
+**Design.** Four architectures, identical in everything except the memory
+mechanism: reward structure, 10×10 world, 100-step episodes, seeds 0–9,
+analysis dimension 256, PPO settings, evaluation protocol, and analysis
+pipeline. They span a memory ladder:
+
+| architecture | memory mechanism | internal state at step t |
+|---|---|---|
+| MLP (`k=8` frame stack) | none (finite window) | pure function of the last 8 observations |
+| SSM (gated diagonal LRU) | compressed, **linear** dynamics | eigenvalue-controlled linear state |
+| RNN (paper's ReLU RNN) | compressed, **nonlinear** dynamics | chaotic nonlinear recurrence |
+| Transformer (2-layer causal) | none evolving; re-attends history | pre-head residual recomputed per step |
+
+Each was trained as a full 20-pair experiment (`mouse-run-run-1` for the RNN,
+`mouse-run-run-2-{mlp,ssm,transformer}` for the rest): 20,000 PPO updates ×
+40 episodes × 100 steps per pair, **20/20 units completed on the first
+attempt** for every architecture, zero failed/interrupted attempts.
+
+---
+
+## 1. Behavior: the architectures converged to different strategies
+
+Chaser capability against a standardized random explorer (100 episodes ×
+100 steps), and the explorer's learned escape (collisions it suffers against a
+random chaser — lower is better escape):
+
+| architecture | chaser collisions ↑ | chaser partner-in-vision | explorer collisions-suffered ↓ |
+|---|---:|---:|---:|
+| RNN | 8.98 ± 5.70 | 0.61 | 0.57 |
+| MLP | 26.93 ± 4.00 | 0.93 | 0.92 |
+| SSM | 35.92 ± 3.56 | 0.94 | 0.45 |
+| Transformer | 41.31 ± 3.29 | 0.96 | 0.28 |
+
+All four learned role-specific social behavior (non-social controls collide
+2.7–3.4× regardless). But the **RNN converged to a qualitatively different
+strategy**: a low-vision "stealth" chaser (61% partner-in-vision, 9 collisions)
+that tracks the explorer from memory, whereas MLP/SSM/Transformer converged to
+high-vision "pursuit" chasers (93–96% vision, 27–41 collisions) that stay
+glued to the explorer. This behavioral fork is the first sign the architectures
+are not learning the same thing — and it turns out to drive the representation
+difference.
+
+The raw capability ranking (Transformer > SSM > MLP > RNN) is **not** the
+headline. A chaser that collides 41×/episode by never leaving the explorer's
+side is behaviorally strong but representationally shallow. The scientific
+question is about the *internal representation*, below.
+
+## 2. Event decoding replicates across architectures (C2)
+
+Linear-classifier balanced accuracy for social events decoded from single-agent
+hidden states (5-fold CV, shuffled controls ≈ 0.50 throughout):
+
+| architecture | collision | partner escape | partner approach |
+|---|---:|---:|---:|
+| RNN | 0.983 | 0.888 | 0.886 |
+| MLP | 0.914 | 0.780 | 0.778 |
+| SSM | 0.850 | 0.768 | 0.733 |
+| Transformer | 0.847 | 0.761 | 0.759 |
+
+Every architecture encodes collision and partner approach/escape above chance
+in social agents — the paper's C2 claim is **architecture-robust**. Decoding
+accuracy tracks the memory ladder (RNN > MLP ≈ SSM > Transformer), but the
+qualitative result (social events are linearly represented) holds for all.
+
+## 3. Shared dimensions: a methodological correction, then the key result (C3)
+
+The paper's headline is C3: social agents develop *shared* neural dimensions
+across the two independently trained networks (PLSC). Measuring this across
+architectures exposed a confound that the analysis had to fix.
+
+**The raw cross-agent correlation is contaminated by time-locked scaffolds.**
+The Transformer's `top_dim_correlation` was **0.957 in the non-social control**
+— higher than its social value, and higher than any architecture's social
+score. There is no partner input in the non-social task, so this cannot be
+genuine sharing. The cause: the Transformer's positional embedding makes every
+timestep a deterministic function of absolute position t; both agents' networks
+carry this identical positional scaffold, and it dominates the top singular
+dimension. The RNN and MLP have no such clean scaffold (their "clocks" are noisy
+or absent), so their raw non-social correlation stays near zero — which is
+exactly why the raw statistic *looked* fine until the Transformer exposed it.
+
+**Fix (standard in neuroscience): remove the evoked/time-locked component.**
+Subtract the across-episode mean at each timestep — the response identical
+across episodes at a given t (positional embeddings, hidden-state clocks, mean
+approach dynamics) — and run PLSC on the episode-specific residual, where
+genuine interaction lives. This removes every architecture's deterministic
+temporal scaffold uniformly, making the effect size comparable.
+
+**Aggregate residualized top-dim correlation (interaction-driven sharing):**
+
+| architecture | social | non-social | separation |
+|---|---:|---:|---:|
+| RNN | 0.688 ± 0.23 | 0.067 | **0.62** |
+| MLP (floor) | 0.679 ± 0.06 | 0.103 | 0.58 |
+| SSM | 0.642 ± 0.12 | 0.315 | 0.33 |
+| Transformer | 0.438 ± 0.08 | 0.356 | **0.08** |
+
+The MLP is the **common-input floor**: being memoryless, any cross-agent sharing
+it shows is by construction reactive (both nets responding to the same visible
+game state). That its social sharing (0.679) nearly matches the RNN's aggregate
+(0.688) is the crux — the aggregate mean *hides* the real difference. The
+Transformer's social/non-social separation collapses to 0.08 once its positional
+scaffold is removed: **its apparent sharing was mostly the scaffold**.
+
+**The discriminating view: sharing vs behavioral coupling (vision).** The MLP
+can only share when both agents see each other. So the question is whether an
+architecture achieves shared representation *at low mutual vision* — which
+requires genuine internal (memory/prediction) coupling, not common input.
+Residualized social `top_r`, split by chaser partner-in-vision:
+
+| architecture | low vision (<40%) | high vision (≥40%) |
+|---|---:|---:|
+| **RNN** | **0.652 (n=7)** | 0.947 (n=1) |
+| MLP | — (0 units) | 0.679 (n=10) |
+| SSM | — (0 units) | 0.642 (n=9) |
+| Transformer | — (0 units) | 0.438 (n=10) |
+
+This is the result. **Only the RNN develops shared representation without the
+agents seeing each other** — 7 of its 8 analyzable social units sit below 40%
+mutual vision yet reach 0.65 residualized correlation. Every social unit of the
+other three architectures lives at high vision; their sharing is vision-gated,
+i.e. common-input-like. MLP *cannot* do otherwise (no memory). SSM and
+Transformer *have* memory mechanisms but did not use them to build low-vision
+coordination — they converged to the reactive, high-vision solution.
+
+## 4. Representational geometry: the RNN is the outlier (CKA)
+
+Feeding the **same** game-state observation stream through every architecture's
+social chaser and comparing representations with linear CKA (rotation/scaling
+invariant):
+
+|  | RNN | MLP | SSM | Transformer |
+|---|---:|---:|---:|---:|
+| RNN | 1.00 | 0.68 | 0.64 | 0.67 |
+| MLP | 0.68 | 1.00 | 0.71 | 0.78 |
+| SSM | 0.64 | 0.71 | 1.00 | 0.76 |
+| Transformer | 0.67 | 0.78 | 0.76 | 1.00 |
+
+The **RNN is the representational outlier** (0.64–0.68 with everything else),
+while MLP/SSM/Transformer cluster together (0.71–0.78). A third independent
+method agrees with behavior and PLSC: the RNN encodes the task differently; the
+other three converge to a more mutually similar (reactive) geometry.
+
+## 5. Degeneracy is itself an architecture property
+
+The paper excludes "degenerate" episodes where agents get stuck. The
+**SSM is markedly more prone to this**: 1/10 social and 2/10 non-social units
+were fully degenerate (too few valid episodes to analyze), and several more had
+only 3–8 of 25 episodes valid — versus **zero fully-degenerate units for RNN,
+MLP, and Transformer**. The linear recurrence appears to collapse toward
+fixed-point / stuck behavior more readily — a concrete failure mode of the
+linear-dynamics inductive bias on this task.
+
+---
+
+## Answer to the question
+
+**Different architectures do not learn the same internal representations.**
+Three independent methods converge:
+
+1. **Behavior** — the RNN alone converges to a low-vision, memory-based stealth
+   strategy; the others to high-vision reactive pursuit.
+2. **Shared dimensions (residualized, vision-conditioned)** — the RNN alone
+   develops genuine interaction-driven sharing when the agents *cannot see each
+   other*; the others' sharing is vision-gated and, for the Transformer, mostly
+   a positional scaffold once corrected.
+3. **CKA** — the RNN is the representational outlier; the other three cluster.
+
+The paper's hallmark phenomenon — genuine internal shared dynamics between
+independently trained agents — is **not architecture-universal**. It emerged
+cleanly only in the vanilla RNN. The memoryless MLP cannot produce it (its
+sharing is the common-input floor). The SSM and Transformer, despite having the
+capacity for memory, did not: their inductive biases steered them to a reactive,
+high-vision solution whose apparent sharing is largely explained by behavioral
+and input coupling, plus (Transformer) a deterministic temporal scaffold.
+
+## Honest confounds and limitations
+
+- **Strategy mediates representation.** The architectures converged to different
+  behavioral strategies, so the representation difference is partly *mediated* by
+  strategy, not a direct architecture→representation effect. But that is itself a
+  finding: the architecture's inductive bias shapes *which* solution is found,
+  and only the RNN's solution has genuine internal shared dynamics. Isolating a
+  direct effect would require conditioning on matched behavior (e.g. training all
+  architectures to the same vision regime), which these runs do not do.
+- **Not parameter-matched.** RNN 119k, SSM 382k, MLP 477k, Transformer 1.24M
+  params (analysis dim held at 256). The RNN's distinctness is not a
+  capacity-advantage artifact — it is the *smallest* model.
+- **Time-mean subtraction** removes any genuinely time-locked coordination along
+  with the scaffold; for random-initial-position episodes the interaction timing
+  is episode-specific, so this should be conservative, but it is an assumption.
+- **PLSC significant-dimension counts saturate** at these sample sizes (MLP hits
+  256/256 in both conditions); the effect size (residualized top correlation),
+  not the count, is the discriminating statistic and is what this report uses.
+- The **perturbation (C5)** and neural-action-space (C4) analyses are not yet run
+  across architectures; C5 in particular (does removing shared dimensions impair
+  social behavior) would test causality of the RNN's shared dimensions.
+
+## Artifacts
+
+- Per-experiment behavior/eval/rollouts/reports: `runs/{experiment}/`,
+  `runs/reports/{experiment}/`.
+- Neural analyses: `runs/analyses/{experiment}/neural_summary.json`.
+- Cross-architecture: `runs/analyses/cross_arch_plsc_residualized.json`,
+  `cross_arch_perunit.json`, `cross_arch_cka.json`.
+- Figures: `runs/analyses/figures/{sharing_vs_vision,cross_arch_cka}.png`.
+- Analysis code: `mouse_run_run/analysis.py`,
+  `scripts/analysis/neural_analyses.py`.

derived/reports/MANIFEST.json

@@ -0,0 +1,71 @@
+{
+  "analysis_script": "/root/mouse-run-run-perf/scripts/analysis/summarize_experiment.py",
+  "generated_at": "2026-07-02T11:51:02.815710+00:00",
+  "input_hashes": {
+    "evaluations.jsonl": "265ed5550a98149da5384ac6cbcd551db5b121176f7046a4ca58d89531a77cc7",
+    "exclusions.jsonl": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+    "rollouts.jsonl": "140147549967098f1a78dadab8e5888eb41a37389c094f79b623d22d7f0bd74a",
+    "runs.jsonl": "bfab762e5dd3673a8e08fa06635ebebd07bc02783f016fc26e56fee41756e2d8"
+  },
+  "input_tables": {
+    "evaluations.jsonl": "runs/tables/mouse-run-run-2-mlp/evaluations.jsonl",
+    "exclusions.jsonl": "runs/tables/mouse-run-run-2-mlp/exclusions.jsonl",
+    "rollouts.jsonl": "runs/tables/mouse-run-run-2-mlp/rollouts.jsonl",
+    "runs.jsonl": "runs/tables/mouse-run-run-2-mlp/runs.jsonl"
+  },
+  "output_hashes": {
+    "evaluation_comparison": "f884d82cdc4bac267de1ab48120986784960a6448d8c7f4502cc3b48c4369735",
+    "training_curves": "f79e51dd791199e40a8057983a80cb9405bc95f186a2e5ee61eae94544f41bac"
+  },
+  "outputs": {
+    "figures": [
+      "runs/reports/mouse-run-run-2-mlp/figures/training_curves.png",
+      "runs/reports/mouse-run-run-2-mlp/figures/evaluation_comparison.png"
+    ],
+    "report": "runs/reports/mouse-run-run-2-mlp/report.md",
+    "summary": "runs/reports/mouse-run-run-2-mlp/summary.json"
+  },
+  "provenance": {
+    "command": "scripts/analysis/summarize_experiment.py runs/tables/mouse-run-run-2-mlp",
+    "created_at": "2026-07-02T11:51:02.816299+00:00",
+    "dependencies": {
+      "huggingface_hub": "1.21.0",
+      "numpy": "2.5.0",
+      "safetensors": "0.8.0",
+      "tensorboard": "2.21.0",
+      "torch": "2.12.1",
+      "triton": "3.7.1"
+    },
+    "executable": "/root/mouse-run-run-perf/.venv/bin/python3",
+    "git": {
+      "branch": "main",
+      "diff_sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+      "dirty": false,
+      "sha": "6282c650d281b31d646d537dd3f57c968806a423",
+      "short_sha": "6282c65",
+      "status_short": ""
+    },
+    "platform": "Linux-6.8.0-117-generic-x86_64-with-glibc2.39",
+    "python": "3.14.5 (main, Jun  2 2026, 22:25:17) [Clang 22.1.3 ]",
+    "schema_version": 1,
+    "torch": {
+      "cuda_available": true,
+      "cuda_device_count": 1,
+      "cuda_devices": [
+        {
+          "index": 0,
+          "major": 8,
+          "minor": 9,
+          "multi_processor_count": 66,
+          "name": "NVIDIA GeForce RTX 4070 Ti SUPER",
+          "total_memory_bytes": 16717578240
+        }
+      ],
+      "cuda_version": "13.0",
+      "cudnn_version": 92000,
+      "mps_available": false,
+      "version": "2.12.1+cu130"
+    }
+  },
+  "schema_version": 2
+}

derived/reports/report.md

@@ -0,0 +1,88 @@
+# MARL Experiment Summary
+
+Generated: `2026-07-02T11:51:02.813569+00:00`
+Tables: `runs/tables/mouse-run-run-2-mlp`
+
+## Valid Pairs
+
+| Task | Valid pairs |
+| --- | ---: |
+| `non_social` | 10 |
+| `social` | 10 |
+
+## Training Metrics (final update, mean ± std across pairs)
+
+| Metric | `non_social` (n=10) | `social` (n=10) |
+| --- | ---: | ---: |
+| collisions per episode | 1.69 ± 0.547 | 20.353 ± 1.679 |
+| chaser return | 6.472 ± 0.343 | 21.552 ± 1.797 |
+| explorer return | 56.474 ± 10.32 | 21.788 ± 4.747 |
+| chaser partner vision | 0 ± 0 | 0.936 ± 0.013 |
+| explorer partner vision | 0 ± 0 | 0.936 ± 0.013 |
+| chaser new fields | 82.878 ± 1.721 | 49.655 ± 2.498 |
+| explorer new fields | 71.42 ± 6.901 | 58.828 ± 3.202 |
+| final distance | 4.834 ± 0.443 | 1.67 ± 0.112 |
+| policy loss | -0.000661 ± 0.000258 | -0.004 ± 0.000856 |
+| value loss | 3.652 ± 0.263 | 4.878 ± 0.277 |
+| entropy | 0.057 ± 0.031 | 0.151 ± 0.02 |
+| approx kl | 0.000278 ± 0.000228 | 0.00088 ± 0.000945 |
+| chaser grad norm | 0.078 ± 0.023 | 0.41 ± 0.212 |
+| explorer grad norm | 0.584 ± 0.277 | 0.549 ± 0.25 |
+| max param abs | 1.452 ± 0.159 | 1.278 ± 0.164 |
+
+## Random-Opponent Evaluation
+
+### Opponent: `random_chaser`
+
+| Metric | `non_social` (n=10) | `social` (n=10) |
+| --- | ---: | ---: |
+| collisions per episode | 2.307 ± 0.275 | 0.924 ± 0.308 |
+| chaser return | -3.304 ± 0.037 | -2.234 ± 0.333 |
+| explorer return | 55.076 ± 9.319 | 27.815 ± 6.397 |
+| chaser partner vision | 0 ± 0 | 0.241 ± 0.028 |
+| explorer partner vision | 0 ± 0 | 0.241 ± 0.028 |
+| chaser new fields | 33.775 ± 0.186 | 33.883 ± 0.057 |
+| explorer new fields | 70.527 ± 6.243 | 52.399 ± 4.267 |
+| final distance | 5.038 ± 0.312 | 5.773 ± 0.251 |
+| average distance | 5.106 ± 0.068 | 5.833 ± 0.193 |
+| degenerate fraction | 0.888 ± 0.031 | 0.892 ± 0.013 |
+
+### Opponent: `random_explorer`
+
+| Metric | `non_social` (n=10) | `social` (n=10) |
+| --- | ---: | ---: |
+| collisions per episode | 2.681 ± 0.243 | 26.935 ± 4.003 |
+| chaser return | 5.923 ± 0.544 | 25.334 ± 4.144 |
+| explorer return | -1.283 ± 0.15 | -28.943 ± 4.167 |
+| chaser partner vision | 0 ± 0 | 0.932 ± 0.017 |
+| explorer partner vision | 0 ± 0 | 0.932 ± 0.017 |
+| chaser new fields | 80.372 ± 2.756 | 31.427 ± 1.323 |
+| explorer new fields | 32.887 ± 0.084 | 26.529 ± 1.279 |
+| final distance | 4.955 ± 0.275 | 1.715 ± 0.099 |
+| average distance | 5.141 ± 0.061 | 1.921 ± 0.113 |
+| degenerate fraction | 0.845 ± 0.014 | 0.95 ± 0.02 |
+
+Failed evaluations: 0
+
+## Figures
+
+### training curves
+
+![training_curves](figures/training_curves.png)
+
+Training curves, mean ± std across 20 valid pairs.
+
+### evaluation comparison
+
+![evaluation_comparison](figures/evaluation_comparison.png)
+
+Evaluation metrics over 40 successful evaluations, grouped by opponent mode and task.
+
+## Degenerate Episode Exclusions
+
+| Excluded episodes | Episodes with records | Records |
+| ---: | ---: | ---: |
+| 0 | 0 | 0 |
+
+## Failed Attempts
+- count: 0

derived/reports/summary.json

@@ -0,0 +1,544 @@
+{
+  "evaluation_metrics": {
+    "failed_count": 0,
+    "non_social": {
+      "random_chaser": {
+        "average_distance": {
+          "max": 5.199133396148682,
+          "mean": 5.105685377120972,
+          "min": 4.98706579208374,
+          "n": 10,
+          "std": 0.06848503924382211
+        },
+        "chaser_new_fields": {
+          "max": 33.97999954223633,
+          "mean": 33.77499961853027,
+          "min": 33.43000030517578,
+          "n": 10,
+          "std": 0.18618548859771447
+        },
+        "chaser_partner_vision": {
+          "max": 0.0,
+          "mean": 0.0,
+          "min": 0.0,
+          "n": 10,
+          "std": 0.0
+        },
+        "chaser_return": {
+          "max": -3.2539987564086914,
+          "mean": -3.3035988092422484,
+          "min": -3.3739986419677734,
+          "n": 10,
+          "std": 0.037113861963604215
+        },
+        "collisions_per_episode": {
+          "max": 2.919999837875366,
+          "mean": 2.306999957561493,
+          "min": 1.8600000143051147,
+          "n": 10,
+          "std": 0.27513812098917756
+        },
+        "degenerate_fraction": {
+          "max": 0.97,
+          "mean": 0.8880000000000001,
+          "min": 0.85,
+          "n": 10,
+          "std": 0.030594117081556706
+        },
+        "explorer_new_fields": {
+          "max": 80.68999481201172,
+          "mean": 70.52699851989746,
+          "min": 60.68000030517578,
+          "n": 10,
+          "std": 6.243129016079577
+        },
+        "explorer_partner_vision": {
+          "max": 0.0,
+          "mean": 0.0,
+          "min": 0.0,
+          "n": 10,
+          "std": 0.0
+        },
+        "explorer_return": {
+          "max": 70.2699966430664,
+          "mean": 55.076498413085936,
+          "min": 40.404998779296875,
+          "n": 10,
+          "std": 9.31878213892601
+        },
+        "final_distance": {
+          "max": 5.588272571563721,
+          "mean": 5.037810468673706,
+          "min": 4.463276386260986,
+          "n": 10,
+          "std": 0.311963295971499
+        },
+        "n": 10
+      },
+      "random_explorer": {
+        "average_distance": {
+          "max": 5.262848854064941,
+          "mean": 5.140882539749145,
+          "min": 5.038296699523926,
+          "n": 10,
+          "std": 0.0614441857456308
+        },
+        "chaser_new_fields": {
+          "max": 86.43999481201172,
+          "mean": 80.3719985961914,
+          "min": 76.8699951171875,
+          "n": 10,
+          "std": 2.755952134734959
+        },
+        "chaser_partner_vision": {
+          "max": 0.0,
+          "mean": 0.0,
+          "min": 0.0,
+          "n": 10,
+          "std": 0.0
+        },
+        "chaser_return": {
+          "max": 7.12199592590332,
+          "mean": 5.922596454620361,
+          "min": 5.221996784210205,
+          "n": 10,
+          "std": 0.5440438394983717
+        },
+        "collisions_per_episode": {
+          "max": 3.049999952316284,
+          "mean": 2.68099992275238,
+          "min": 2.309999942779541,
+          "n": 10,
+          "std": 0.24341115086841367
+        },
+        "degenerate_fraction": {
+          "max": 0.87,
+          "mean": 0.845,
+          "min": 0.83,
+          "n": 10,
+          "std": 0.013601470508735456
+        },
+        "explorer_new_fields": {
+          "max": 33.040000915527344,
+          "mean": 32.88699913024902,
+          "min": 32.790000915527344,
+          "n": 10,
+          "std": 0.08391110303705251
+        },
+        "explorer_partner_vision": {
+          "max": 0.0,
+          "mean": 0.0,
+          "min": 0.0,
+          "n": 10,
+          "std": 0.0
+        },
+        "explorer_return": {
+          "max": -0.9799999594688416,
+          "mean": -1.282999974489212,
+          "min": -1.5199999809265137,
+          "n": 10,
+          "std": 0.14951924109474032
+        },
+        "final_distance": {
+          "max": 5.392402648925781,
+          "mean": 4.9553193092346195,
+          "min": 4.301638603210449,
+          "n": 10,
+          "std": 0.2753013389573719
+        },
+        "n": 10
+      }
+    },
+    "social": {
+      "random_chaser": {
+        "average_distance": {
+          "max": 6.055323123931885,
+          "mean": 5.832867050170899,
+          "min": 5.418818473815918,
+          "n": 10,
+          "std": 0.1926549744871535
+        },
+        "chaser_new_fields": {
+          "max": 33.95000076293945,
+          "mean": 33.88299980163574,
+          "min": 33.73999786376953,
+          "n": 10,
+          "std": 0.05728056247021595
+        },
+        "chaser_partner_vision": {
+          "max": 0.2913999855518341,
+          "mean": 0.24062999188899994,
+          "min": 0.204599991440773,
+          "n": 10,
+          "std": 0.028305234515883754
+        },
+        "chaser_return": {
+          "max": -1.4439992904663086,
+          "mean": -2.2341991543769835,
+          "min": -2.5089991092681885,
+          "n": 10,
+          "std": 0.3334365212524313
+        },
+        "collisions_per_episode": {
+          "max": 1.659999966621399,
+          "mean": 0.9239999949932098,
+          "min": 0.6699999570846558,
+          "n": 10,
+          "std": 0.3079350544139646
+        },
+        "degenerate_fraction": {
+          "max": 0.91,
+          "mean": 0.892,
+          "min": 0.87,
+          "n": 10,
+          "std": 0.013266499161421611
+        },
+        "explorer_new_fields": {
+          "max": 57.55999755859375,
+          "mean": 52.3989990234375,
+          "min": 43.37999725341797,
+          "n": 10,
+          "std": 4.266789586753233
+        },
+        "explorer_partner_vision": {
+          "max": 0.2913999855518341,
+          "mean": 0.24062999188899994,
+          "min": 0.204599991440773,
+          "n": 10,
+          "std": 0.028305234515883754
+        },
+        "explorer_return": {
+          "max": 35.744998931884766,
+          "mean": 27.815499019622802,
+          "min": 14.429999351501465,
+          "n": 10,
+          "std": 6.396563521591426
+        },
+        "final_distance": {
+          "max": 6.047131061553955,
+          "mean": 5.773358011245728,
+          "min": 5.25391149520874,
+          "n": 10,
+          "std": 0.2512269355601858
+        },
+        "n": 10
+      },
+      "random_explorer": {
+        "average_distance": {
+          "max": 2.13496994972229,
+          "mean": 1.921411967277527,
+          "min": 1.7290464639663696,
+          "n": 10,
+          "std": 0.11278872658258984
+        },
+        "chaser_new_fields": {
+          "max": 33.89999771118164,
+          "mean": 31.42699909210205,
+          "min": 28.920000076293945,
+          "n": 10,
+          "std": 1.3234575205271881
+        },
+        "chaser_partner_vision": {
+          "max": 0.9477999806404114,
+          "mean": 0.9316199898719788,
+          "min": 0.8927000164985657,
+          "n": 10,
+          "std": 0.017225497745840646
+        },
+        "chaser_return": {
+          "max": 36.003997802734375,
+          "mean": 25.334495544433594,
+          "min": 20.284996032714844,
+          "n": 10,
+          "std": 4.144235438399579
+        },
+        "collisions_per_episode": {
+          "max": 37.13999938964844,
+          "mean": 26.934999465942383,
+          "min": 21.850000381469727,
+          "n": 10,
+          "std": 4.003446545522908
+        },
+        "degenerate_fraction": {
+          "max": 0.98,
+          "mean": 0.95,
+          "min": 0.92,
+          "n": 10,
+          "std": 0.020493901531919184
+        },
+        "explorer_new_fields": {
+          "max": 28.149999618530273,
+          "mean": 26.528999519348144,
+          "min": 23.78999900817871,
+          "n": 10,
+          "std": 1.2791048421652784
+        },
+        "explorer_partner_vision": {
+          "max": 0.9477999806404114,
+          "mean": 0.9316199898719788,
+          "min": 0.8927000164985657,
+          "n": 10,
+          "std": 0.017225497745840646
+        },
+        "explorer_return": {
+          "max": -23.260000228881836,
+          "mean": -28.943499755859374,
+          "min": -39.125,
+          "n": 10,
+          "std": 4.167097957623246
+        },
+        "final_distance": {
+          "max": 1.8503373861312866,
+          "mean": 1.7153441309928894,
+          "min": 1.5262326002120972,
+          "n": 10,
+          "std": 0.09933532411070188
+        },
+        "n": 10
+      }
+    }
+  },
+  "failed_attempts": [],
+  "figures": [
+    {
+      "caption": "Training curves, mean \u00b1 std across 20 valid pairs.",
+      "name": "training_curves",
+      "path": "runs/reports/mouse-run-run-2-mlp/figures/training_curves.png"
+    },
+    {
+      "caption": "Evaluation metrics over 40 successful evaluations, grouped by opponent mode and task.",
+      "name": "evaluation_comparison",
+      "path": "runs/reports/mouse-run-run-2-mlp/figures/evaluation_comparison.png"
+    }
+  ],
+  "generated_at": "2026-07-02T11:51:02.813569+00:00",
+  "rollout_exclusions": {
+    "episodes_with_exclusion_records": 0,
+    "excluded_degenerate_episodes": 0,
+    "records": 0
+  },
+  "schema_version": 2,
+  "tables_root": "runs/tables/mouse-run-run-2-mlp",
+  "training_metrics": {
+    "non_social": {
+      "approx_kl": {
+        "max": 0.0007710583850932262,
+        "mean": 0.00027836359963407043,
+        "min": 3.927184380664306e-06,
+        "n": 10,
+        "std": 0.00022781666240450588
+      },
+      "chaser_grad_norm": {
+        "max": 0.11076641827821732,
+        "mean": 0.07840223964303732,
+        "min": 0.04017048981040716,
+        "n": 10,
+        "std": 0.023169039813528615
+      },
+      "chaser_new_fields": {
+        "max": 87.07500457763672,
+        "mean": 82.87750015258788,
+        "min": 80.4749984741211,
+        "n": 10,
+        "std": 1.7213541677267494
+      },
+      "chaser_partner_vision": {
+        "max": 0.0,
+        "mean": 0.0,
+        "min": 0.0,
+        "n": 10,
+        "std": 0.0
+      },
+      "chaser_return": {
+        "max": 7.300001621246338,
+        "mean": 6.471501016616822,
+        "min": 6.015000820159912,
+        "n": 10,
+        "std": 0.34321339840677706
+      },
+      "collisions_per_episode": {
+        "max": 2.7250001430511475,
+        "mean": 1.690000021457672,
+        "min": 0.9750000238418579,
+        "n": 10,
+        "std": 0.5471745917052043
+      },
+      "entropy": {
+        "max": 0.13418130949139595,
+        "mean": 0.05714834872633219,
+        "min": 0.02094564540311694,
+        "n": 10,
+        "std": 0.030692080850040007
+      },
+      "explorer_grad_norm": {
+        "max": 1.1887003779411316,
+        "mean": 0.583899623900652,
+        "min": 0.2844117097556591,
+        "n": 10,
+        "std": 0.27718636321364126
+      },
+      "explorer_new_fields": {
+        "max": 82.75,
+        "mean": 71.42000007629395,
+        "min": 60.57500076293945,
+        "n": 10,
+        "std": 6.900640552089833
+      },
+      "explorer_partner_vision": {
+        "max": 0.0,
+        "mean": 0.0,
+        "min": 0.0,
+        "n": 10,
+        "std": 0.0
+      },
+      "explorer_return": {
+        "max": 73.30000305175781,
+        "mean": 56.473750305175784,
+        "min": 40.07500076293945,
+        "n": 10,
+        "std": 10.319657580882787
+      },
+      "final_distance": {
+        "max": 5.683727741241455,
+        "mean": 4.834419059753418,
+        "min": 4.144174098968506,
+        "n": 10,
+        "std": 0.4431673642962756
+      },
+      "max_param_abs": {
+        "max": 1.7384926080703735,
+        "mean": 1.452017331123352,
+        "min": 1.2401388883590698,
+        "n": 10,
+        "std": 0.15908309762880601
+      },
+      "n": 10,
+      "policy_loss": {
+        "max": -0.0002411644566997495,
+        "mean": -0.000661178331059098,
+        "min": -0.0011177757633866747,
+        "n": 10,
+        "std": 0.00025768936399840804
+      },
+      "value_loss": {
+        "max": 4.014994144439697,
+        "mean": 3.651918852329254,
+        "min": 3.322889506816864,
+        "n": 10,
+        "std": 0.2633674319523318
+      }
+    },
+    "social": {
+      "approx_kl": {
+        "max": 0.002805363851033249,
+        "mean": 0.00087992062623089,
+        "min": -0.0002899035942244277,
+        "n": 10,
+        "std": 0.0009453947863305684
+      },
+      "chaser_grad_norm": {
+        "max": 0.9264960587024689,
+        "mean": 0.41003721952438354,
+        "min": 0.188689436763525,
+        "n": 10,
+        "std": 0.2119915449421527
+      },
+      "chaser_new_fields": {
+        "max": 52.92499923706055,
+        "mean": 49.65500030517578,
+        "min": 45.57500076293945,
+        "n": 10,
+        "std": 2.4984694774842433
+      },
+      "chaser_partner_vision": {
+        "max": 0.9552500247955322,
+        "mean": 0.9360250353813171,
+        "min": 0.9205000400543213,
+        "n": 10,
+        "std": 0.012655263263627304
+      },
+      "chaser_return": {
+        "max": 24.285001754760742,
+        "mean": 21.551751327514648,
+        "min": 18.48750114440918,
+        "n": 10,
+        "std": 1.796639770827262
+      },
+      "collisions_per_episode": {
+        "max": 23.200000762939453,
+        "mean": 20.35250053405762,
+        "min": 17.625,
+        "n": 10,
+        "std": 1.6785577964252976
+      },
+      "entropy": {
+        "max": 0.19220001995563507,
+        "mean": 0.15101509019732476,
+        "min": 0.12773804366588593,
+        "n": 10,
+        "std": 0.020022485274942464
+      },
+      "explorer_grad_norm": {
+        "max": 1.1695344895124435,
+        "mean": 0.5494304157793521,
+        "min": 0.23243309557437897,
+        "n": 10,
+        "std": 0.2504391396265317
+      },
+      "explorer_new_fields": {
+        "max": 65.55000305175781,
+        "mean": 58.827500915527345,
+        "min": 52.775001525878906,
+        "n": 10,
+        "std": 3.201532040899581
+      },
+      "explorer_partner_vision": {
+        "max": 0.9552500247955322,
+        "mean": 0.9360250353813171,
+        "min": 0.9205000400543213,
+        "n": 10,
+        "std": 0.012655263263627304
+      },
+      "explorer_return": {
+        "max": 31.037500381469727,
+        "mean": 21.787500286102294,
+        "min": 12.71250057220459,
+        "n": 10,
+        "std": 4.746788332300563
+      },
+      "final_distance": {
+        "max": 1.8305482864379883,
+        "mean": 1.670472300052643,
+        "min": 1.4895528554916382,
+        "n": 10,
+        "std": 0.1115461884593616
+      },
+      "max_param_abs": {
+        "max": 1.6354095935821533,
+        "mean": 1.2778974533081056,
+        "min": 1.0662928819656372,
+        "n": 10,
+        "std": 0.16398531671135025
+      },
+      "n": 10,
+      "policy_loss": {
+        "max": -0.0027591757117768623,
+        "mean": -0.00390199398393738,
+        "min": -0.005259661337355226,
+        "n": 10,
+        "std": 0.0008555915976877569
+      },
+      "value_loss": {
+        "max": 5.285007953643799,
+        "mean": 4.877697789669037,
+        "min": 4.4600828886032104,
+        "n": 10,
+        "std": 0.27699674303611843
+      }
+    }
+  },
+  "valid_pairs_by_task": {
+    "non_social": 10,
+    "social": 10
+  }
+}

derived/tables/MANIFEST.json

@@ -0,0 +1,67 @@
+{
+  "experiment_root": "runs/mouse-run-run-2-mlp",
+  "generated_at": "2026-07-02T11:51:00.748915+00:00",
+  "output_hashes": {
+    "evaluations": "265ed5550a98149da5384ac6cbcd551db5b121176f7046a4ca58d89531a77cc7",
+    "exclusions": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+    "rollouts": "140147549967098f1a78dadab8e5888eb41a37389c094f79b623d22d7f0bd74a",
+    "runs": "bfab762e5dd3673a8e08fa06635ebebd07bc02783f016fc26e56fee41756e2d8"
+  },
+  "outputs": {
+    "evaluations": "runs/tables/mouse-run-run-2-mlp/evaluations.jsonl",
+    "exclusions": "runs/tables/mouse-run-run-2-mlp/exclusions.jsonl",
+    "rollouts": "runs/tables/mouse-run-run-2-mlp/rollouts.jsonl",
+    "runs": "runs/tables/mouse-run-run-2-mlp/runs.jsonl"
+  },
+  "panel_path": "runs/tables/mouse-run-run-2-mlp/panels.json",
+  "provenance": {
+    "command": "scripts/transform/build_tables.py runs/mouse-run-run-2-mlp",
+    "created_at": "2026-07-02T11:51:00.749193+00:00",
+    "dependencies": {
+      "huggingface_hub": "1.21.0",
+      "numpy": "2.5.0",
+      "safetensors": "0.8.0",
+      "tensorboard": "2.21.0",
+      "torch": "2.12.1",
+      "triton": "3.7.1"
+    },
+    "executable": "/root/mouse-run-run-perf/.venv/bin/python3",
+    "git": {
+      "branch": "main",
+      "diff_sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+      "dirty": false,
+      "sha": "6282c650d281b31d646d537dd3f57c968806a423",
+      "short_sha": "6282c65",
+      "status_short": ""
+    },
+    "platform": "Linux-6.8.0-117-generic-x86_64-with-glibc2.39",
+    "python": "3.14.5 (main, Jun  2 2026, 22:25:17) [Clang 22.1.3 ]",
+    "schema_version": 1,
+    "torch": {
+      "cuda_available": true,
+      "cuda_device_count": 1,
+      "cuda_devices": [
+        {
+          "index": 0,
+          "major": 8,
+          "minor": 9,
+          "multi_processor_count": 66,
+          "name": "NVIDIA GeForce RTX 4070 Ti SUPER",
+          "total_memory_bytes": 16717578240
+        }
+      ],
+      "cuda_version": "13.0",
+      "cudnn_version": 92000,
+      "mps_available": false,
+      "version": "2.12.1+cu130"
+    }
+  },
+  "row_counts": {
+    "evaluations": 40,
+    "exclusions": 0,
+    "rollouts": 20,
+    "runs": 20
+  },
+  "schema_version": 1,
+  "transform_script": "/root/mouse-run-run-perf/scripts/transform/build_tables.py"
+}

derived/tables/panels.json

@@ -0,0 +1,28 @@
+{
+  "panels": {
+    "paper_neural_behavior_non_degenerate_v1": {
+      "exclude_by": "exclusions.jsonl reason=paper_degenerate_episode_v1",
+      "table": "rollouts.jsonl",
+      "unit": "checkpoint rollout episode"
+    },
+    "paper_random_opponent_valid_v1": {
+      "include": {
+        "evaluation_protocol": "paper_random_opponent_v1",
+        "status": "ok"
+      },
+      "table": "evaluations.jsonl",
+      "unit": "checkpoint x opponent_mode"
+    },
+    "primary_valid_runs_v1": {
+      "include": {
+        "checkpoint_ok": true,
+        "healthy": true,
+        "is_latest_successful": true,
+        "status": "completed"
+      },
+      "table": "runs.jsonl",
+      "unit": "task x seed latest successful attempt"
+    }
+  },
+  "schema_version": 1
+}

derived/tables/rollouts.jsonl

@@ -0,0 +1,20 @@
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:48.710464+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0000__attempt_01__checkpoints__latest__9e74dbbabd.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01", "schema_version": 2, "seed": 0, "source_line": 1, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0000"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:49.261015+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0001__attempt_01__checkpoints__latest__6aac5a2c12.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01", "schema_version": 2, "seed": 0, "source_line": 2, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0001"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:49.810154+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0002__attempt_01__checkpoints__latest__33f1a8c09e.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01", "schema_version": 2, "seed": 0, "source_line": 3, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0002"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:50.371542+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0003__attempt_01__checkpoints__latest__ddb70257e3.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01", "schema_version": 2, "seed": 0, "source_line": 4, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0003"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:50.918340+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0004__attempt_01__checkpoints__latest__8dc5f10a59.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01", "schema_version": 2, "seed": 0, "source_line": 5, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0004"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:51.471824+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0005__attempt_01__checkpoints__latest__c46a11793f.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01", "schema_version": 2, "seed": 0, "source_line": 6, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0005"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:52.025242+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0006__attempt_01__checkpoints__latest__c29253393a.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01", "schema_version": 2, "seed": 0, "source_line": 7, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0006"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:52.569627+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0007__attempt_01__checkpoints__latest__a64ce3bd0c.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01", "schema_version": 2, "seed": 0, "source_line": 8, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0007"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:53.116716+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0008__attempt_01__checkpoints__latest__3f8cb5f7d2.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01", "schema_version": 2, "seed": 0, "source_line": 9, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0008"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:53.672368+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__non_social__seed_0009__attempt_01__checkpoints__latest__e145c2cab5.safetensors", "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01", "schema_version": 2, "seed": 0, "source_line": 10, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "non_social", "unit_id": "non_social/seed_0009"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:54.273478+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0000__attempt_01__checkpoints__latest__f2865fcba9.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01", "schema_version": 2, "seed": 0, "source_line": 11, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0000"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:54.873436+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0001__attempt_01__checkpoints__latest__61848b4a30.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01", "schema_version": 2, "seed": 0, "source_line": 12, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0001"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:55.461961+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0002__attempt_01__checkpoints__latest__77d0837868.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01", "schema_version": 2, "seed": 0, "source_line": 13, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0002"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:56.048411+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0003__attempt_01__checkpoints__latest__3d4e5109f0.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01", "schema_version": 2, "seed": 0, "source_line": 14, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0003"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:56.645581+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0004__attempt_01__checkpoints__latest__4399b2519b.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01", "schema_version": 2, "seed": 0, "source_line": 15, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0004"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:57.229171+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0005__attempt_01__checkpoints__latest__7ddcdb43f2.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01", "schema_version": 2, "seed": 0, "source_line": 16, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0005"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:57.814519+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0006__attempt_01__checkpoints__latest__b87e617200.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01", "schema_version": 2, "seed": 0, "source_line": 17, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0006"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:58.414469+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0007__attempt_01__checkpoints__latest__b494d062b3.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01", "schema_version": 2, "seed": 0, "source_line": 18, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0007"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:58.996792+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0008__attempt_01__checkpoints__latest__c502c04b75.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01", "schema_version": 2, "seed": 0, "source_line": 19, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0008"}
+{"attempt_id": "attempt_01", "batch_size": 25, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "command": "/root/mouse-run-run-perf/.venv/bin/collect-paper-marl-rollouts runs/mouse-run-run-2-mlp --output-root runs/mouse-run-run-2-mlp/paper_rollouts --device cuda", "created_at": "2026-07-02T11:50:59.591392+00:00", "degenerate_episode_count": 0, "degenerate_threshold_fraction": 0.01, "deterministic": false, "device": "cuda", "episode_count": 25, "episodes": 25, "error": null, "experiment_id": "mouse-run-run-2-mlp", "max_steps": 500, "output": "runs/mouse-run-run-2-mlp/paper_rollouts/mouse-run-run-2-mlp__social__seed_0009__attempt_01__checkpoints__latest__b92a4c3ad0.safetensors", "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01", "schema_version": 2, "seed": 0, "source_line": 20, "source_path": "runs/mouse-run-run-2-mlp/paper_rollouts/paper_rollout_records.jsonl", "status": "ok", "task": "social", "unit_id": "social/seed_0009"}

derived/tables/runs.jsonl

@@ -0,0 +1,20 @@
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01", "seed": 0, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7602.132536457037, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0002434901476933704, "chaser_grad_norm": 0.07690595369786024, "chaser_new_fields": 83.0999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.540000915527344, "chaser_subspace_norm": 75.278076171875, "collisions_per_episode": 1.25, "entropy": 0.03620169032365084, "explorer_grad_norm": 0.2844117097556591, "explorer_new_fields": 75.5250015258789, "explorer_partner_vision": 0.0, "explorer_return": 62.76250076293945, "explorer_subspace_norm": 28.751102447509766, "final_distance": 5.306502342224121, "max_param_abs": 1.7384926080703735, "policy_loss": -0.0007488939012390006, "value_loss": 3.4187601804733276}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01", "schema_version": 1, "seed": 0, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0000", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01", "seed": 1, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7618.575776674086, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0007710583850932262, "chaser_grad_norm": 0.11076641827821732, "chaser_new_fields": 83.25, "chaser_partner_vision": 0.0, "chaser_return": 6.595000743865967, "chaser_subspace_norm": 107.82763671875, "collisions_per_episode": 0.9750000238418579, "entropy": 0.06922571919858456, "explorer_grad_norm": 0.3988708183169365, "explorer_new_fields": 73.5, "explorer_partner_vision": 0.0, "explorer_return": 59.76250076293945, "explorer_subspace_norm": 27.074764251708984, "final_distance": 4.443612575531006, "max_param_abs": 1.3260666131973267, "policy_loss": -0.0006433645908252572, "value_loss": 3.4030256271362305}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01", "schema_version": 1, "seed": 1, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0001", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01", "seed": 2, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7622.409991315915, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.00015983218477721994, "chaser_grad_norm": 0.06889200676232576, "chaser_new_fields": 84.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.7600016593933105, "chaser_subspace_norm": 98.9588623046875, "collisions_per_episode": 2.075000047683716, "entropy": 0.02094564540311694, "explorer_grad_norm": 1.1887003779411316, "explorer_new_fields": 82.75, "explorer_partner_vision": 0.0, "explorer_return": 73.30000305175781, "explorer_subspace_norm": 38.70703887939453, "final_distance": 4.667576789855957, "max_param_abs": 1.5199339389801025, "policy_loss": -0.0006238509232616707, "value_loss": 3.509905695915222}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01", "schema_version": 1, "seed": 2, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0002", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01", "seed": 3, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7631.6458840209525, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0002841827333717878, "chaser_grad_norm": 0.04017048981040716, "chaser_new_fields": 87.07500457763672, "chaser_partner_vision": 0.0, "chaser_return": 7.300001621246338, "chaser_subspace_norm": 112.96394348144531, "collisions_per_episode": 1.725000023841858, "entropy": 0.024153232108801603, "explorer_grad_norm": 0.8466866165399551, "explorer_new_fields": 75.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 62.125, "explorer_subspace_norm": 36.111541748046875, "final_distance": 4.144174098968506, "max_param_abs": 1.6294050216674805, "policy_loss": -0.0005663236785924397, "value_loss": 3.4905534386634827}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01", "schema_version": 1, "seed": 3, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0003", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01", "seed": 4, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7640.98172001401, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.00037036755928841064, "chaser_grad_norm": 0.08058446925133467, "chaser_new_fields": 82.17500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.330001354217529, "chaser_subspace_norm": 92.44800567626953, "collisions_per_episode": 1.8250000476837158, "entropy": 0.04575740918517113, "explorer_grad_norm": 0.3279200941324234, "explorer_new_fields": 77.9749984741211, "explorer_partner_vision": 0.0, "explorer_return": 66.0999984741211, "explorer_subspace_norm": 29.62271499633789, "final_distance": 5.308696269989014, "max_param_abs": 1.3726928234100342, "policy_loss": -0.0006299224167634065, "value_loss": 3.322889506816864}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01", "schema_version": 1, "seed": 4, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0004", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01", "seed": 5, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7562.313855465036, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.00022429173202174373, "chaser_grad_norm": 0.04249562695622444, "chaser_new_fields": 80.4749984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.015000820159912, "chaser_subspace_norm": 93.73216247558594, "collisions_per_episode": 1.024999976158142, "entropy": 0.05529557354748249, "explorer_grad_norm": 0.39152009785175323, "explorer_new_fields": 73.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 59.42499923706055, "explorer_subspace_norm": 117.07512664794922, "final_distance": 4.730400562286377, "max_param_abs": 1.606895923614502, "policy_loss": -0.0002411644566997495, "value_loss": 3.523293375968933}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01", "schema_version": 1, "seed": 5, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0005", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01", "seed": 6, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7567.199341987027, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 8.635803430578903e-06, "chaser_grad_norm": 0.10996681079268456, "chaser_new_fields": 82.375, "chaser_partner_vision": 0.0, "chaser_return": 6.390000820159912, "chaser_subspace_norm": 73.99776458740234, "collisions_per_episode": 1.149999976158142, "entropy": 0.05159799661487341, "explorer_grad_norm": 0.7565127648413181, "explorer_new_fields": 61.025001525878906, "explorer_partner_vision": 0.0, "explorer_return": 41.23749923706055, "explorer_subspace_norm": 46.546077728271484, "final_distance": 4.668344020843506, "max_param_abs": 1.2401388883590698, "policy_loss": -0.0002673804239634592, "value_loss": 3.8410890698432922}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01", "schema_version": 1, "seed": 6, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0006", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01", "seed": 7, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7575.734206710942, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0005693047922270544, "chaser_grad_norm": 0.08362068422138691, "chaser_new_fields": 82.0250015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.3000006675720215, "chaser_subspace_norm": 109.71160888671875, "collisions_per_episode": 2.174999952316284, "entropy": 0.05865261983126402, "explorer_grad_norm": 0.6362885981798172, "explorer_new_fields": 60.57500076293945, "explorer_partner_vision": 0.0, "explorer_return": 40.07500076293945, "explorer_subspace_norm": 35.61513900756836, "final_distance": 4.870059967041016, "max_param_abs": 1.4202895164489746, "policy_loss": -0.0008471746835780891, "value_loss": 4.014994144439697}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01", "schema_version": 1, "seed": 7, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0007", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01", "seed": 8, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7577.744137497037, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 3.927184380664306e-06, "chaser_grad_norm": 0.07202051393687725, "chaser_new_fields": 81.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.145000457763672, "chaser_subspace_norm": 132.7053680419922, "collisions_per_episode": 1.975000023841858, "entropy": 0.13418130949139595, "explorer_grad_norm": 0.6797178164124489, "explorer_new_fields": 68.7249984741211, "explorer_partner_vision": 0.0, "explorer_return": 52.48749923706055, "explorer_subspace_norm": 61.86991500854492, "final_distance": 4.521096229553223, "max_param_abs": 1.4186153411865234, "policy_loss": -0.0009259324722812323, "value_loss": 4.002953112125397}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01", "schema_version": 1, "seed": 8, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0008", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "none", "task": "non_social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01", "seed": 9, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7570.7650183150545, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.00014854547405664803, "chaser_grad_norm": 0.09859942272305489, "chaser_new_fields": 82.5999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.340001106262207, "chaser_subspace_norm": 98.81183624267578, "collisions_per_episode": 2.7250001430511475, "entropy": 0.07547229155898094, "explorer_grad_norm": 0.32836734503507614, "explorer_new_fields": 65.67500305175781, "explorer_partner_vision": 0.0, "explorer_return": 47.462501525878906, "explorer_subspace_norm": 26.826025009155273, "final_distance": 5.683727741241455, "max_param_abs": 1.2476426362991333, "policy_loss": -0.0011177757633866747, "value_loss": 3.991724371910095}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01", "schema_version": 1, "seed": 9, "status": "completed", "task": "non_social", "total_updates": 20000, "unit_id": "non_social/seed_0009", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01", "seed": 0, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7633.100595657015, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.002805363851033249, "chaser_grad_norm": 0.188689436763525, "chaser_new_fields": 52.92499923706055, "chaser_partner_vision": 0.9490000605583191, "chaser_return": 23.260000228881836, "chaser_subspace_norm": 46.499481201171875, "collisions_per_episode": 21.450000762939453, "entropy": 0.1341038942337036, "explorer_grad_norm": 1.1695344895124435, "explorer_new_fields": 65.55000305175781, "explorer_partner_vision": 0.9490000605583191, "explorer_return": 31.037500381469727, "explorer_subspace_norm": 105.87859344482422, "final_distance": 1.655848741531372, "max_param_abs": 1.2022045850753784, "policy_loss": -0.003802475623324142, "value_loss": 4.788339614868164}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01", "schema_version": 1, "seed": 0, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0000", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01", "seed": 1, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7644.564348316053, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0005562261293015069, "chaser_grad_norm": 0.6423081681132317, "chaser_new_fields": 46.20000076293945, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 23.4325008392334, "chaser_subspace_norm": 57.10129928588867, "collisions_per_episode": 22.625, "entropy": 0.14692536741495132, "explorer_grad_norm": 0.5817638859152794, "explorer_new_fields": 52.775001525878906, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 12.71250057220459, "explorer_subspace_norm": 122.46495819091797, "final_distance": 1.791115403175354, "max_param_abs": 1.3355721235275269, "policy_loss": -0.004476253691127141, "value_loss": 5.1513988971710205}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01", "schema_version": 1, "seed": 1, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0001", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01", "seed": 2, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7640.485689831898, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0014266868641836794, "chaser_grad_norm": 0.20290611311793327, "chaser_new_fields": 49.29999923706055, "chaser_partner_vision": 0.9287500381469727, "chaser_return": 18.48750114440918, "chaser_subspace_norm": 48.7114143371582, "collisions_per_episode": 17.625, "entropy": 0.18204283714294434, "explorer_grad_norm": 0.6118505597114563, "explorer_new_fields": 60.95000076293945, "explorer_partner_vision": 0.9287500381469727, "explorer_return": 26.837499618530273, "explorer_subspace_norm": 104.41957092285156, "final_distance": 1.6572967767715454, "max_param_abs": 1.1167881488800049, "policy_loss": -0.005259661337355226, "value_loss": 5.285007953643799}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01", "schema_version": 1, "seed": 2, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0002", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01", "seed": 3, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7642.0857080380665, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.000801136315581541, "chaser_grad_norm": 0.38031933456659317, "chaser_new_fields": 52.5, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 21.880001068115234, "chaser_subspace_norm": 52.41925811767578, "collisions_per_episode": 20.200000762939453, "entropy": 0.12773804366588593, "explorer_grad_norm": 0.23243309557437897, "explorer_new_fields": 59.29999923706055, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 22.887500762939453, "explorer_subspace_norm": 119.99281311035156, "final_distance": 1.4895528554916382, "max_param_abs": 1.4659936428070068, "policy_loss": -0.0027591757117768623, "value_loss": 5.230631113052368}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01", "schema_version": 1, "seed": 3, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0003", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01", "seed": 4, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7640.117992118001, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.000236154562574864, "chaser_grad_norm": 0.38722849637269974, "chaser_new_fields": 45.57500076293945, "chaser_partner_vision": 0.9225000143051147, "chaser_return": 19.41750144958496, "chaser_subspace_norm": 58.249393463134766, "collisions_per_episode": 19.125, "entropy": 0.19220001995563507, "explorer_grad_norm": 0.3825383335351944, "explorer_new_fields": 56.85000228881836, "explorer_partner_vision": 0.9225000143051147, "explorer_return": 19.30000114440918, "explorer_subspace_norm": 90.54417419433594, "final_distance": 1.830224633216858, "max_param_abs": 1.0662928819656372, "policy_loss": -0.004796283046713867, "value_loss": 5.130034923553467}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01", "schema_version": 1, "seed": 4, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0004", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01", "seed": 5, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7556.8758046310395, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0005827252787078407, "chaser_grad_norm": 0.3728112056851387, "chaser_new_fields": 51.47500228881836, "chaser_partner_vision": 0.95250004529953, "chaser_return": 20.117502212524414, "chaser_subspace_norm": 50.84516143798828, "collisions_per_episode": 18.575000762939453, "entropy": 0.13566195219755173, "explorer_grad_norm": 0.6062372475862503, "explorer_new_fields": 58.57500076293945, "explorer_partner_vision": 0.95250004529953, "explorer_return": 20.774999618530273, "explorer_subspace_norm": 104.22787475585938, "final_distance": 1.5808099508285522, "max_param_abs": 1.3590010404586792, "policy_loss": -0.003127762281341262, "value_loss": 4.4600828886032104}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01", "schema_version": 1, "seed": 5, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0005", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01", "seed": 6, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7558.8218611279735, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.00025015939701367174, "chaser_grad_norm": 0.31481391191482544, "chaser_new_fields": 49.900001525878906, "chaser_partner_vision": 0.9252500534057617, "chaser_return": 21.020002365112305, "chaser_subspace_norm": 49.21955108642578, "collisions_per_episode": 19.80000114440918, "entropy": 0.1560382954776287, "explorer_grad_norm": 0.6951030343770981, "explorer_new_fields": 59.42499923706055, "explorer_partner_vision": 0.9252500534057617, "explorer_return": 22.48750114440918, "explorer_subspace_norm": 113.72798919677734, "final_distance": 1.6350990533828735, "max_param_abs": 1.6354095935821533, "policy_loss": -0.0035889741121788443, "value_loss": 4.686129331588745}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01", "schema_version": 1, "seed": 6, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0006", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01", "seed": 7, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7587.008876845008, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.00013653355120357702, "chaser_grad_norm": 0.9264960587024689, "chaser_new_fields": 46.92499923706055, "chaser_partner_vision": 0.9470000267028809, "chaser_return": 24.285001754760742, "chaser_subspace_norm": 67.47857666015625, "collisions_per_episode": 23.200000762939453, "entropy": 0.1407805271446705, "explorer_grad_norm": 0.5267750471830368, "explorer_new_fields": 55.900001525878906, "explorer_partner_vision": 0.9470000267028809, "explorer_return": 17.149999618530273, "explorer_subspace_norm": 89.71617889404297, "final_distance": 1.689988374710083, "max_param_abs": 1.2198518514633179, "policy_loss": -0.003363337086513596, "value_loss": 4.716352343559265}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01", "schema_version": 1, "seed": 7, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0007", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01", "seed": 8, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7584.383489511092, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": 0.0022941239069333985, "chaser_grad_norm": 0.43133843690156937, "chaser_new_fields": 50.125, "chaser_partner_vision": 0.9205000400543213, "chaser_return": 20.852500915527344, "chaser_subspace_norm": 52.702186584472656, "collisions_per_episode": 19.67500114440918, "entropy": 0.15483620017766953, "explorer_grad_norm": 0.3490636646747589, "explorer_new_fields": 60.525001525878906, "explorer_partner_vision": 0.9205000400543213, "explorer_return": 23.0, "explorer_subspace_norm": 101.96681213378906, "final_distance": 1.8305482864379883, "max_param_abs": 1.1802449226379395, "policy_loss": -0.00493374926503698, "value_loss": 4.697651386260986}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01", "schema_version": 1, "seed": 8, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0008", "update": 20000}
+{"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_health": {"failures": [], "format_ok": true, "max_abs": 80000.0, "nonfinite_tensor_count": 0, "ok": true, "tensor_count": 66}, "checkpoint_ok": true, "config": {"architecture": "mlp", "attempt_id": "attempt_01", "batch_size": 40, "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "checkpoint_every": 1000, "checkpoint_every_seconds": 1800.0, "clip_epsilon": 0.2, "cost_per_hour": null, "cuda_tf32": true, "device": "cuda", "entropy_coef": 0.01, "env": {"grid_size": 10, "max_steps": 100, "partner_visibility": "partial", "task": "social", "vision_radius": 3}, "experiment_id": "mouse-run-run-2-mlp", "finite_guard": true, "fused_agent_rollout": false, "gae_lambda": 0.95, "gamma": 0.99, "grad_clip": 1.0, "hidden_size": 256, "learning_rate": 0.0003, "log_every": 500, "metrics_path": null, "ppo_epochs": 4, "recurrent_l2_coef": 0.3, "resume_from": null, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01", "seed": 9, "status_every_seconds": 1800.0, "status_path": null, "subspace_metric_period": 10, "tensorboard_dir": null, "triton_env_step": true, "updates": 20000, "value_clip": 10.0, "value_coef": 0.5}, "elapsed_seconds": 7585.64040600392, "environment_steps": 80000000, "error": null, "eta_seconds": 0.0, "experiment": "mouse-run-run-2-mlp", "healthy": true, "is_latest_successful": true, "metrics": {"approx_kl": -0.0002899035942244277, "chaser_grad_norm": 0.2534610331058502, "chaser_new_fields": 51.625, "chaser_partner_vision": 0.9552500247955322, "chaser_return": 22.76500129699707, "chaser_subspace_norm": 57.69636535644531, "collisions_per_episode": 21.25, "entropy": 0.1398237645626068, "explorer_grad_norm": 0.3390047997236252, "explorer_new_fields": 58.42499923706055, "explorer_partner_vision": 0.9552500247955322, "explorer_return": 21.6875, "explorer_subspace_norm": 105.33492279052734, "final_distance": 1.5442389249801636, "max_param_abs": 1.1976157426834106, "policy_loss": -0.002912267684005876, "value_loss": 4.631349444389343}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01", "schema_version": 1, "seed": 9, "status": "completed", "task": "social", "total_updates": 20000, "unit_id": "social/seed_0009", "update": 20000}

launch_gate.json

@@ -0,0 +1,204 @@
+{
+  "checks": [
+    {
+      "detail": "experiments/paper_marl_2026/SPEC.md",
+      "name": "experiment_spec",
+      "ok": true
+    },
+    {
+      "detail": {
+        "missing": []
+      },
+      "name": "config_schema",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": 20000,
+        "expected": 20000
+      },
+      "name": "primary_updates",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": 40,
+        "expected": 40
+      },
+      "name": "primary_batch_size",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": 100,
+        "expected": 100
+      },
+      "name": "primary_max_steps",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": 10,
+        "expected": 10
+      },
+      "name": "primary_seeds",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": 0,
+        "expected": 0
+      },
+      "name": "primary_seed_start",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": "both",
+        "expected": "both"
+      },
+      "name": "primary_tasks",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": "paper_text",
+        "expected": "paper_text"
+      },
+      "name": "primary_preset",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": "cuda",
+        "expected": "cuda"
+      },
+      "name": "primary_device",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": true,
+        "expected": true
+      },
+      "name": "primary_finite_guard",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": true,
+        "expected": true
+      },
+      "name": "primary_triton_env_step",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": false,
+        "expected": false
+      },
+      "name": "primary_fused_agent_rollout",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": 10,
+        "expected": 10
+      },
+      "name": "primary_subspace_metric_period",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual": 2
+      },
+      "name": "max_attempts",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual_seconds": 1800.0
+      },
+      "name": "observability_period",
+      "ok": true
+    },
+    {
+      "detail": {
+        "actual_hours": 24.0
+      },
+      "name": "attempt_timeout",
+      "ok": true
+    },
+    {
+      "detail": {
+        "cuda_available": true,
+        "device": "cuda"
+      },
+      "name": "device_available",
+      "ok": true
+    },
+    {
+      "detail": {
+        "current_dirty": false,
+        "current_git_sha": "6282c650d281b31d646d537dd3f57c968806a423",
+        "path": "runs/mouse-run-run-2-mlp/triton_equivalence.json",
+        "record_dirty": false,
+        "record_git_sha": "6282c650d281b31d646d537dd3f57c968806a423",
+        "sha_matches": true,
+        "status": "ok"
+      },
+      "name": "triton_equivalence",
+      "ok": true
+    }
+  ],
+  "config": "experiments/paper_marl_2026/configs/arch_mlp.json",
+  "config_sha256": "4d12552c6b3cc4801827d0193b91d5ab945da76650f3691d3086729e2cc0c723",
+  "created_at": "2026-07-02T07:36:15.459261+00:00",
+  "experiment_spec": "experiments/paper_marl_2026/SPEC.md",
+  "experiment_spec_sha256": "e878bd03a06735c93c8caab4fc769ce63f026a538dfccbc87b588d432e48761e",
+  "provenance": {
+    "command": "scripts/run/launch_gate.py --config experiments/paper_marl_2026/configs/arch_mlp.json --triton-equivalence runs/mouse-run-run-2-mlp/triton_equivalence.json",
+    "created_at": "2026-07-02T07:36:15.459353+00:00",
+    "dependencies": {
+      "huggingface_hub": "1.21.0",
+      "numpy": "2.5.0",
+      "safetensors": "0.8.0",
+      "tensorboard": "2.21.0",
+      "torch": "2.12.1",
+      "triton": "3.7.1"
+    },
+    "executable": "/root/mouse-run-run-perf/.venv/bin/python3",
+    "git": {
+      "branch": "main",
+      "diff_sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+      "dirty": false,
+      "sha": "6282c650d281b31d646d537dd3f57c968806a423",
+      "short_sha": "6282c65",
+      "status_short": ""
+    },
+    "platform": "Linux-6.8.0-117-generic-x86_64-with-glibc2.39",
+    "python": "3.14.5 (main, Jun  2 2026, 22:25:17) [Clang 22.1.3 ]",
+    "schema_version": 1,
+    "torch": {
+      "cuda_available": true,
+      "cuda_device_count": 1,
+      "cuda_devices": [
+        {
+          "index": 0,
+          "major": 8,
+          "minor": 9,
+          "multi_processor_count": 66,
+          "name": "NVIDIA GeForce RTX 4070 Ti SUPER",
+          "total_memory_bytes": 16717578240
+        }
+      ],
+      "cuda_version": "13.0",
+      "cudnn_version": 92000,
+      "mps_available": false,
+      "version": "2.12.1+cu130"
+    }
+  },
+  "schema_version": 1,
+  "status": "ok",
+  "triton_equivalence": "runs/mouse-run-run-2-mlp/triton_equivalence.json"
+}

logs/non_social_seed_0000_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.521858+00:00 task=non_social seed=0 attempt=1 ===
+update=0001 collisions=1.70 chaser_return=-2.91 explorer_return=0.10 vision=(0.000,0.000) new_fields=(35.8,33.7) distance=5.47 loss=(-0.000,3.074) kl=-0.0000
+update=0500 collisions=3.38 chaser_return=4.70 explorer_return=49.94 vision=(0.000,0.000) new_fields=(74.2,67.3) distance=4.96 loss=(-0.002,4.558) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=1.90 chaser_return=5.81 explorer_return=50.42 vision=(0.000,0.000) new_fields=(79.5,67.3) distance=5.55 loss=(-0.002,4.399) kl=0.0007
+update=1500 collisions=3.05 chaser_return=6.08 explorer_return=53.65 vision=(0.000,0.000) new_fields=(80.8,69.6) distance=5.11 loss=(-0.001,3.947) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=2.12 chaser_return=5.82 explorer_return=50.42 vision=(0.000,0.000) new_fields=(79.7,67.3) distance=4.40 loss=(-0.001,4.047) kl=0.0008
+update=2500 collisions=2.30 chaser_return=5.18 explorer_return=55.11 vision=(0.000,0.000) new_fields=(76.7,70.8) distance=5.46 loss=(-0.001,3.958) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=2.38 chaser_return=5.36 explorer_return=50.58 vision=(0.000,0.000) new_fields=(77.6,67.6) distance=5.60 loss=(-0.002,3.992) kl=0.0004
+update=3500 collisions=1.90 chaser_return=6.02 explorer_return=54.36 vision=(0.000,0.000) new_fields=(80.7,70.0) distance=4.94 loss=(-0.000,3.771) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=1.65 chaser_return=5.88 explorer_return=57.17 vision=(0.000,0.000) new_fields=(79.8,71.9) distance=5.85 loss=(-0.001,3.668) kl=0.0001
+update=4500 collisions=1.62 chaser_return=6.29 explorer_return=54.14 vision=(0.000,0.000) new_fields=(81.8,69.8) distance=5.53 loss=(-0.001,3.734) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=2.15 chaser_return=5.38 explorer_return=59.09 vision=(0.000,0.000) new_fields=(77.4,73.2) distance=5.58 loss=(-0.000,3.802) kl=0.0008
+update=5500 collisions=1.62 chaser_return=5.69 explorer_return=58.53 vision=(0.000,0.000) new_fields=(78.8,72.8) distance=5.20 loss=(-0.001,3.841) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=1.85 chaser_return=5.56 explorer_return=57.70 vision=(0.000,0.000) new_fields=(78.3,72.4) distance=4.69 loss=(-0.001,3.698) kl=0.0004
+update=6500 collisions=1.73 chaser_return=6.44 explorer_return=58.60 vision=(0.000,0.000) new_fields=(82.6,72.9) distance=5.05 loss=(-0.001,3.696) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=1.60 chaser_return=6.43 explorer_return=65.50 vision=(0.000,0.000) new_fields=(82.6,77.5) distance=4.87 loss=(-0.000,3.672) kl=-0.0000
+update=7500 collisions=1.77 chaser_return=5.95 explorer_return=65.05 vision=(0.000,0.000) new_fields=(80.2,77.2) distance=5.50 loss=(-0.001,3.700) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=1.65 chaser_return=5.95 explorer_return=63.25 vision=(0.000,0.000) new_fields=(80.3,76.0) distance=5.81 loss=(-0.000,3.573) kl=0.0000
+update=8500 collisions=1.93 chaser_return=6.19 explorer_return=64.56 vision=(0.000,0.000) new_fields=(81.4,76.7) distance=5.31 loss=(-0.001,3.583) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=1.83 chaser_return=5.74 explorer_return=66.33 vision=(0.000,0.000) new_fields=(79.3,78.0) distance=4.99 loss=(-0.001,3.665) kl=0.0005
+update=9500 collisions=1.83 chaser_return=6.16 explorer_return=64.94 vision=(0.000,0.000) new_fields=(81.3,77.1) distance=4.54 loss=(-0.000,3.484) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=2.25 chaser_return=6.16 explorer_return=65.28 vision=(0.000,0.000) new_fields=(81.4,77.2) distance=5.32 loss=(-0.000,3.523) kl=-0.0002
+update=10500 collisions=2.00 chaser_return=6.42 explorer_return=65.20 vision=(0.000,0.000) new_fields=(82.7,77.1) distance=5.14 loss=(-0.001,3.464) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=1.52 chaser_return=6.52 explorer_return=66.59 vision=(0.000,0.000) new_fields=(83.1,78.2) distance=4.93 loss=(-0.000,3.494) kl=0.0001
+update=11500 collisions=1.40 chaser_return=6.07 explorer_return=69.33 vision=(0.000,0.000) new_fields=(80.8,80.0) distance=5.81 loss=(-0.002,3.456) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=1.65 chaser_return=5.11 explorer_return=66.18 vision=(0.000,0.000) new_fields=(75.9,78.0) distance=5.17 loss=(-0.001,3.763) kl=0.0008
+update=12500 collisions=1.35 chaser_return=6.79 explorer_return=68.12 vision=(0.000,0.000) new_fields=(84.3,79.1) distance=5.73 loss=(-0.001,3.352) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=1.40 chaser_return=6.55 explorer_return=67.64 vision=(0.000,0.000) new_fields=(83.2,78.7) distance=5.54 loss=(-0.001,3.472) kl=0.0005
+update=13500 collisions=1.73 chaser_return=6.42 explorer_return=69.70 vision=(0.000,0.000) new_fields=(82.6,80.3) distance=5.57 loss=(-0.002,3.334) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=1.83 chaser_return=6.42 explorer_return=66.62 vision=(0.000,0.000) new_fields=(82.8,78.2) distance=5.71 loss=(-0.001,3.345) kl=0.0002
+update=14500 collisions=1.70 chaser_return=6.39 explorer_return=68.01 vision=(0.000,0.000) new_fields=(82.6,79.1) distance=5.15 loss=(-0.000,3.397) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=1.40 chaser_return=6.77 explorer_return=67.53 vision=(0.000,0.000) new_fields=(84.5,78.7) distance=5.50 loss=(-0.000,3.233) kl=0.0003
+update=15500 collisions=1.18 chaser_return=6.56 explorer_return=63.85 vision=(0.000,0.000) new_fields=(83.1,76.2) distance=5.34 loss=(-0.001,3.442) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=1.70 chaser_return=6.42 explorer_return=69.96 vision=(0.000,0.000) new_fields=(82.7,80.5) distance=5.65 loss=(-0.001,3.293) kl=0.0001
+update=16500 collisions=1.58 chaser_return=6.63 explorer_return=71.50 vision=(0.000,0.000) new_fields=(83.6,81.5) distance=5.41 loss=(-0.000,3.280) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=1.48 chaser_return=6.36 explorer_return=70.04 vision=(0.000,0.000) new_fields=(82.3,80.5) distance=5.11 loss=(-0.000,3.309) kl=0.0002
+update=17500 collisions=1.35 chaser_return=6.46 explorer_return=71.80 vision=(0.000,0.000) new_fields=(82.8,81.5) distance=5.24 loss=(-0.000,3.166) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=1.18 chaser_return=6.41 explorer_return=68.84 vision=(0.000,0.000) new_fields=(82.4,79.7) distance=5.66 loss=(-0.000,3.226) kl=0.0005
+update=18500 collisions=1.52 chaser_return=6.22 explorer_return=69.55 vision=(0.000,0.000) new_fields=(81.5,80.0) distance=6.22 loss=(-0.001,3.249) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=1.30 chaser_return=6.22 explorer_return=65.99 vision=(0.000,0.000) new_fields=(81.6,77.7) distance=5.97 loss=(-0.000,3.290) kl=-0.0000
+update=19500 collisions=1.52 chaser_return=6.90 explorer_return=64.11 vision=(0.000,0.000) new_fields=(85.1,76.4) distance=6.13 loss=(-0.001,3.472) kl=0.0001
+update=20000 collisions=1.25 chaser_return=6.54 explorer_return=62.76 vision=(0.000,0.000) new_fields=(83.1,75.5) distance=5.31 loss=(-0.001,3.419) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7602.184716115007, "metrics": {"approx_kl": 0.0002434901476933704, "chaser_grad_norm": 0.07690595369786024, "chaser_new_fields": 83.0999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.540000915527344, "chaser_subspace_norm": 75.278076171875, "collisions_per_episode": 1.25, "entropy": 0.03620169032365084, "explorer_grad_norm": 0.2844117097556591, "explorer_new_fields": 75.5250015258789, "explorer_partner_vision": 0.0, "explorer_return": 62.76250076293945, "explorer_subspace_norm": 28.751102447509766, "final_distance": 5.306502342224121, "max_param_abs": 1.7384926080703735, "policy_loss": -0.0007488939012390006, "value_loss": 3.4187601804733276}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01", "seed": 0, "task": "non_social", "unit_id": "non_social/seed_0000"}
+=== exited 2026-07-02T09:43:08.049146+00:00 returncode=0 ===

logs/non_social_seed_0001_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.523172+00:00 task=non_social seed=1 attempt=1 ===
+update=0001 collisions=1.10 chaser_return=-3.14 explorer_return=4.04 vision=(0.000,0.000) new_fields=(34.5,36.2) distance=5.17 loss=(-0.001,2.900) kl=-0.0000
+update=0500 collisions=1.05 chaser_return=5.33 explorer_return=48.62 vision=(0.000,0.000) new_fields=(76.9,66.1) distance=5.66 loss=(-0.002,4.656) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=1.08 chaser_return=5.80 explorer_return=50.28 vision=(0.000,0.000) new_fields=(79.2,67.0) distance=5.04 loss=(-0.002,4.247) kl=0.0003
+update=1500 collisions=1.25 chaser_return=6.27 explorer_return=47.73 vision=(0.000,0.000) new_fields=(81.6,65.5) distance=4.56 loss=(-0.001,4.188) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=1.30 chaser_return=6.26 explorer_return=50.95 vision=(0.000,0.000) new_fields=(81.6,67.7) distance=5.20 loss=(-0.000,3.860) kl=0.0006
+update=2500 collisions=1.30 chaser_return=6.16 explorer_return=52.45 vision=(0.000,0.000) new_fields=(81.2,68.6) distance=4.25 loss=(-0.001,4.090) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=1.80 chaser_return=6.49 explorer_return=56.39 vision=(0.000,0.000) new_fields=(82.9,71.2) distance=4.91 loss=(-0.000,3.652) kl=0.0005
+update=3500 collisions=1.38 chaser_return=6.53 explorer_return=56.05 vision=(0.000,0.000) new_fields=(83.1,71.0) distance=4.81 loss=(-0.001,3.655) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=1.73 chaser_return=6.30 explorer_return=57.29 vision=(0.000,0.000) new_fields=(81.9,71.9) distance=4.88 loss=(-0.001,3.659) kl=-0.0001
+update=4500 collisions=1.48 chaser_return=6.14 explorer_return=52.26 vision=(0.000,0.000) new_fields=(81.1,68.6) distance=4.94 loss=(-0.000,3.886) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=1.18 chaser_return=5.84 explorer_return=53.01 vision=(0.000,0.000) new_fields=(79.7,69.0) distance=5.93 loss=(-0.000,3.810) kl=0.0002
+update=5500 collisions=1.68 chaser_return=5.90 explorer_return=56.95 vision=(0.000,0.000) new_fields=(79.9,71.6) distance=4.47 loss=(-0.001,3.734) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=1.25 chaser_return=5.76 explorer_return=55.41 vision=(0.000,0.000) new_fields=(79.1,70.6) distance=5.37 loss=(-0.000,3.706) kl=0.0001
+update=6500 collisions=1.40 chaser_return=5.33 explorer_return=60.40 vision=(0.000,0.000) new_fields=(76.9,74.0) distance=5.48 loss=(-0.000,3.598) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=1.18 chaser_return=5.19 explorer_return=62.61 vision=(0.000,0.000) new_fields=(76.4,75.4) distance=6.20 loss=(-0.001,4.010) kl=0.0001
+update=7500 collisions=1.02 chaser_return=6.66 explorer_return=63.06 vision=(0.000,0.000) new_fields=(83.7,75.6) distance=5.51 loss=(-0.000,3.763) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=1.05 chaser_return=6.45 explorer_return=63.70 vision=(0.000,0.000) new_fields=(82.4,76.2) distance=5.56 loss=(-0.001,3.566) kl=0.0005
+update=8500 collisions=1.00 chaser_return=6.19 explorer_return=65.16 vision=(0.000,0.000) new_fields=(81.2,77.2) distance=5.27 loss=(-0.001,3.467) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=0.85 chaser_return=6.68 explorer_return=65.76 vision=(0.000,0.000) new_fields=(83.7,77.4) distance=5.15 loss=(-0.000,3.594) kl=0.0001
+update=9500 collisions=1.00 chaser_return=4.38 explorer_return=59.46 vision=(0.000,0.000) new_fields=(72.2,73.3) distance=5.29 loss=(-0.000,3.870) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=1.02 chaser_return=6.09 explorer_return=62.31 vision=(0.000,0.000) new_fields=(80.7,75.3) distance=5.23 loss=(-0.001,3.616) kl=0.0004
+update=10500 collisions=1.43 chaser_return=5.69 explorer_return=58.98 vision=(0.000,0.000) new_fields=(78.9,73.2) distance=4.75 loss=(-0.001,3.461) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=2.23 chaser_return=6.07 explorer_return=58.98 vision=(0.000,0.000) new_fields=(80.7,73.1) distance=4.46 loss=(-0.001,3.696) kl=0.0002
+update=11500 collisions=1.40 chaser_return=5.59 explorer_return=65.58 vision=(0.000,0.000) new_fields=(78.5,77.5) distance=5.03 loss=(-0.001,3.508) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=1.60 chaser_return=5.33 explorer_return=54.36 vision=(0.000,0.000) new_fields=(77.0,70.1) distance=5.15 loss=(-0.001,3.632) kl=0.0002
+update=12500 collisions=1.90 chaser_return=5.13 explorer_return=55.60 vision=(0.000,0.000) new_fields=(76.1,70.9) distance=4.85 loss=(-0.001,3.693) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=1.15 chaser_return=5.46 explorer_return=61.19 vision=(0.000,0.000) new_fields=(77.6,74.6) distance=5.08 loss=(-0.001,3.686) kl=0.0008
+update=13500 collisions=1.70 chaser_return=5.43 explorer_return=56.28 vision=(0.000,0.000) new_fields=(77.5,71.2) distance=5.20 loss=(-0.001,3.767) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=2.03 chaser_return=5.36 explorer_return=56.58 vision=(0.000,0.000) new_fields=(77.4,71.5) distance=4.74 loss=(-0.001,3.665) kl=0.0002
+update=14500 collisions=1.93 chaser_return=5.19 explorer_return=51.85 vision=(0.000,0.000) new_fields=(76.5,68.4) distance=5.30 loss=(-0.002,3.699) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=1.77 chaser_return=5.49 explorer_return=58.60 vision=(0.000,0.000) new_fields=(77.9,72.9) distance=5.04 loss=(-0.001,3.715) kl=0.0002
+update=15500 collisions=1.35 chaser_return=5.79 explorer_return=60.29 vision=(0.000,0.000) new_fields=(79.3,73.9) distance=5.41 loss=(-0.001,3.430) kl=0.0015
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=1.35 chaser_return=5.40 explorer_return=57.51 vision=(0.000,0.000) new_fields=(77.3,72.0) distance=5.17 loss=(-0.000,3.629) kl=0.0003
+update=16500 collisions=1.93 chaser_return=5.49 explorer_return=53.28 vision=(0.000,0.000) new_fields=(77.9,69.4) distance=5.08 loss=(-0.001,3.668) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=1.88 chaser_return=5.45 explorer_return=58.00 vision=(0.000,0.000) new_fields=(77.9,72.5) distance=5.44 loss=(-0.001,3.733) kl=0.0002
+update=17500 collisions=1.83 chaser_return=5.95 explorer_return=60.33 vision=(0.000,0.000) new_fields=(80.2,74.2) distance=5.85 loss=(-0.000,3.431) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=1.58 chaser_return=5.58 explorer_return=62.24 vision=(0.000,0.000) new_fields=(78.2,75.2) distance=5.50 loss=(-0.001,3.572) kl=0.0002
+update=18500 collisions=1.48 chaser_return=5.70 explorer_return=51.70 vision=(0.000,0.000) new_fields=(79.0,68.2) distance=4.39 loss=(-0.001,3.885) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=1.50 chaser_return=6.04 explorer_return=58.30 vision=(0.000,0.000) new_fields=(80.8,72.6) distance=4.76 loss=(-0.000,3.455) kl=0.0002
+update=19500 collisions=0.90 chaser_return=6.27 explorer_return=57.10 vision=(0.000,0.000) new_fields=(81.6,71.7) distance=4.83 loss=(-0.000,3.564) kl=0.0006
+update=20000 collisions=0.98 chaser_return=6.60 explorer_return=59.76 vision=(0.000,0.000) new_fields=(83.2,73.5) distance=4.44 loss=(-0.001,3.403) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7618.63883137505, "metrics": {"approx_kl": 0.0007710583850932262, "chaser_grad_norm": 0.11076641827821732, "chaser_new_fields": 83.25, "chaser_partner_vision": 0.0, "chaser_return": 6.595000743865967, "chaser_subspace_norm": 107.82763671875, "collisions_per_episode": 0.9750000238418579, "entropy": 0.06922571919858456, "explorer_grad_norm": 0.3988708183169365, "explorer_new_fields": 73.5, "explorer_partner_vision": 0.0, "explorer_return": 59.76250076293945, "explorer_subspace_norm": 27.074764251708984, "final_distance": 4.443612575531006, "max_param_abs": 1.3260666131973267, "policy_loss": -0.0006433645908252572, "value_loss": 3.4030256271362305}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01", "seed": 1, "task": "non_social", "unit_id": "non_social/seed_0001"}
+=== exited 2026-07-02T09:43:18.367158+00:00 returncode=0 ===

logs/non_social_seed_0002_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.524044+00:00 task=non_social seed=2 attempt=1 ===
+update=0001 collisions=1.80 chaser_return=-3.30 explorer_return=4.97 vision=(0.000,0.000) new_fields=(33.8,36.8) distance=5.51 loss=(-0.001,3.103) kl=0.0000
+update=0500 collisions=2.25 chaser_return=4.99 explorer_return=44.84 vision=(0.000,0.000) new_fields=(75.3,63.8) distance=5.45 loss=(-0.001,4.764) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=2.62 chaser_return=5.26 explorer_return=47.69 vision=(0.000,0.000) new_fields=(76.8,65.6) distance=5.27 loss=(-0.002,4.376) kl=0.0007
+update=1500 collisions=2.58 chaser_return=4.76 explorer_return=51.78 vision=(0.000,0.000) new_fields=(74.4,68.5) distance=5.16 loss=(-0.002,4.042) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=3.25 chaser_return=4.16 explorer_return=53.50 vision=(0.000,0.000) new_fields=(71.5,69.5) distance=5.11 loss=(-0.001,4.174) kl=0.0000
+update=2500 collisions=2.45 chaser_return=4.94 explorer_return=60.44 vision=(0.000,0.000) new_fields=(75.4,74.1) distance=5.53 loss=(-0.001,3.834) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=2.20 chaser_return=5.45 explorer_return=61.75 vision=(0.000,0.000) new_fields=(77.8,75.0) distance=5.20 loss=(-0.001,3.790) kl=-0.0001
+update=3500 collisions=2.40 chaser_return=5.00 explorer_return=55.98 vision=(0.000,0.000) new_fields=(75.6,71.1) distance=5.27 loss=(-0.002,4.077) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=3.20 chaser_return=5.04 explorer_return=56.99 vision=(0.000,0.000) new_fields=(76.1,71.9) distance=5.15 loss=(-0.001,3.932) kl=0.0001
+update=4500 collisions=2.03 chaser_return=4.83 explorer_return=61.15 vision=(0.000,0.000) new_fields=(74.7,74.6) distance=4.90 loss=(-0.001,3.991) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=3.03 chaser_return=5.47 explorer_return=57.48 vision=(0.000,0.000) new_fields=(78.2,72.2) distance=4.81 loss=(-0.001,3.771) kl=0.0002
+update=5500 collisions=2.08 chaser_return=5.41 explorer_return=60.59 vision=(0.000,0.000) new_fields=(77.6,74.3) distance=4.74 loss=(-0.001,3.858) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=2.38 chaser_return=5.44 explorer_return=59.88 vision=(0.000,0.000) new_fields=(77.9,73.6) distance=4.70 loss=(-0.000,4.051) kl=0.0000
+update=6500 collisions=2.33 chaser_return=5.87 explorer_return=68.01 vision=(0.000,0.000) new_fields=(79.9,79.1) distance=5.67 loss=(-0.000,3.768) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=2.62 chaser_return=4.46 explorer_return=69.36 vision=(0.000,0.000) new_fields=(73.0,80.2) distance=5.27 loss=(-0.001,3.815) kl=0.0006
+update=7500 collisions=2.75 chaser_return=5.92 explorer_return=66.74 vision=(0.000,0.000) new_fields=(80.3,78.4) distance=4.29 loss=(-0.001,3.648) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=2.35 chaser_return=5.92 explorer_return=69.29 vision=(0.000,0.000) new_fields=(80.1,79.9) distance=4.07 loss=(-0.001,3.647) kl=0.0003
+update=8500 collisions=2.33 chaser_return=5.90 explorer_return=70.30 vision=(0.000,0.000) new_fields=(80.1,80.6) distance=3.76 loss=(-0.001,3.613) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=1.93 chaser_return=6.26 explorer_return=75.10 vision=(0.000,0.000) new_fields=(81.8,83.8) distance=4.52 loss=(-0.000,3.443) kl=0.0006
+update=9500 collisions=2.40 chaser_return=6.10 explorer_return=74.88 vision=(0.000,0.000) new_fields=(81.0,83.7) distance=4.31 loss=(-0.001,3.433) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=2.12 chaser_return=5.89 explorer_return=73.41 vision=(0.000,0.000) new_fields=(80.0,82.7) distance=4.59 loss=(-0.000,3.523) kl=-0.0001
+update=10500 collisions=2.78 chaser_return=5.69 explorer_return=61.38 vision=(0.000,0.000) new_fields=(79.1,74.8) distance=5.01 loss=(-0.001,3.884) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=2.23 chaser_return=6.36 explorer_return=61.90 vision=(0.000,0.000) new_fields=(82.2,75.0) distance=4.50 loss=(-0.001,3.602) kl=0.0003
+update=11500 collisions=1.43 chaser_return=6.49 explorer_return=71.76 vision=(0.000,0.000) new_fields=(82.9,81.5) distance=4.82 loss=(-0.000,3.341) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=2.80 chaser_return=6.28 explorer_return=64.75 vision=(0.000,0.000) new_fields=(82.1,77.0) distance=4.86 loss=(-0.002,3.599) kl=0.0006
+update=12500 collisions=1.93 chaser_return=6.36 explorer_return=66.93 vision=(0.000,0.000) new_fields=(82.2,78.4) distance=4.70 loss=(-0.000,3.529) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=1.75 chaser_return=6.35 explorer_return=69.21 vision=(0.000,0.000) new_fields=(82.2,79.9) distance=4.77 loss=(-0.001,3.496) kl=0.0005
+update=13500 collisions=1.98 chaser_return=6.57 explorer_return=66.70 vision=(0.000,0.000) new_fields=(83.6,78.1) distance=5.22 loss=(-0.000,3.530) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=2.58 chaser_return=6.49 explorer_return=71.84 vision=(0.000,0.000) new_fields=(83.1,81.9) distance=5.08 loss=(-0.000,3.381) kl=-0.0000
+update=14500 collisions=1.90 chaser_return=6.67 explorer_return=70.08 vision=(0.000,0.000) new_fields=(83.9,80.5) distance=4.94 loss=(-0.000,3.423) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=2.08 chaser_return=6.53 explorer_return=69.03 vision=(0.000,0.000) new_fields=(83.1,79.8) distance=4.93 loss=(-0.001,3.598) kl=-0.0000
+update=15500 collisions=2.20 chaser_return=6.24 explorer_return=70.00 vision=(0.000,0.000) new_fields=(81.9,80.7) distance=4.71 loss=(-0.001,3.464) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=1.50 chaser_return=6.61 explorer_return=70.94 vision=(0.000,0.000) new_fields=(83.4,81.0) distance=5.13 loss=(-0.000,3.484) kl=0.0000
+update=16500 collisions=2.00 chaser_return=6.35 explorer_return=70.15 vision=(0.000,0.000) new_fields=(82.5,80.5) distance=4.24 loss=(-0.000,3.485) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=2.10 chaser_return=6.20 explorer_return=69.21 vision=(0.000,0.000) new_fields=(81.5,79.8) distance=4.89 loss=(-0.001,3.600) kl=0.0006
+update=17500 collisions=1.93 chaser_return=6.54 explorer_return=66.93 vision=(0.000,0.000) new_fields=(83.1,78.3) distance=5.16 loss=(-0.001,3.538) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=1.98 chaser_return=5.89 explorer_return=69.14 vision=(0.000,0.000) new_fields=(80.0,79.8) distance=5.03 loss=(-0.001,3.525) kl=0.0009
+update=18500 collisions=4.85 chaser_return=5.94 explorer_return=66.55 vision=(0.000,0.000) new_fields=(80.2,78.0) distance=4.18 loss=(-0.001,3.637) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=2.40 chaser_return=6.31 explorer_return=69.29 vision=(0.000,0.000) new_fields=(82.1,80.1) distance=4.77 loss=(-0.000,3.486) kl=0.0004
+update=19500 collisions=2.42 chaser_return=6.49 explorer_return=72.51 vision=(0.000,0.000) new_fields=(83.0,82.1) distance=5.00 loss=(-0.000,3.395) kl=0.0001
+update=20000 collisions=2.08 chaser_return=6.76 explorer_return=73.30 vision=(0.000,0.000) new_fields=(84.3,82.8) distance=4.67 loss=(-0.001,3.510) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7622.430775379995, "metrics": {"approx_kl": 0.00015983218477721994, "chaser_grad_norm": 0.06889200676232576, "chaser_new_fields": 84.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.7600016593933105, "chaser_subspace_norm": 98.9588623046875, "collisions_per_episode": 2.075000047683716, "entropy": 0.02094564540311694, "explorer_grad_norm": 1.1887003779411316, "explorer_new_fields": 82.75, "explorer_partner_vision": 0.0, "explorer_return": 73.30000305175781, "explorer_subspace_norm": 38.70703887939453, "final_distance": 4.667576789855957, "max_param_abs": 1.5199339389801025, "policy_loss": -0.0006238509232616707, "value_loss": 3.509905695915222}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01", "seed": 2, "task": "non_social", "unit_id": "non_social/seed_0002"}
+=== exited 2026-07-02T09:43:28.528507+00:00 returncode=0 ===

logs/non_social_seed_0003_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.524887+00:00 task=non_social seed=3 attempt=1 ===
+update=0001 collisions=2.10 chaser_return=-3.79 explorer_return=-0.57 vision=(0.000,0.000) new_fields=(31.4,33.2) distance=4.79 loss=(-0.000,3.267) kl=0.0000
+update=0500 collisions=3.30 chaser_return=5.14 explorer_return=48.55 vision=(0.000,0.000) new_fields=(76.2,66.1) distance=6.01 loss=(-0.002,4.514) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=4.10 chaser_return=5.51 explorer_return=46.67 vision=(0.000,0.000) new_fields=(78.3,65.2) distance=4.67 loss=(-0.002,4.498) kl=0.0004
+update=1500 collisions=2.48 chaser_return=6.05 explorer_return=43.34 vision=(0.000,0.000) new_fields=(80.9,62.7) distance=5.01 loss=(-0.001,4.305) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=2.33 chaser_return=5.94 explorer_return=42.21 vision=(0.000,0.000) new_fields=(80.6,61.9) distance=5.07 loss=(-0.001,4.197) kl=0.0005
+update=2500 collisions=2.38 chaser_return=6.28 explorer_return=43.23 vision=(0.000,0.000) new_fields=(82.1,62.6) distance=5.09 loss=(-0.000,4.095) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=2.33 chaser_return=6.37 explorer_return=43.08 vision=(0.000,0.000) new_fields=(82.6,62.5) distance=5.17 loss=(-0.001,3.895) kl=0.0002
+update=3500 collisions=2.50 chaser_return=6.58 explorer_return=46.08 vision=(0.000,0.000) new_fields=(83.7,64.7) distance=6.12 loss=(-0.001,3.854) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=2.28 chaser_return=6.61 explorer_return=45.48 vision=(0.000,0.000) new_fields=(83.8,64.0) distance=5.39 loss=(-0.000,3.766) kl=-0.0002
+update=4500 collisions=1.93 chaser_return=6.38 explorer_return=44.73 vision=(0.000,0.000) new_fields=(82.7,63.5) distance=5.78 loss=(-0.001,3.812) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=1.73 chaser_return=6.71 explorer_return=43.98 vision=(0.000,0.000) new_fields=(84.2,63.0) distance=6.08 loss=(-0.000,3.856) kl=0.0001
+update=5500 collisions=2.35 chaser_return=6.35 explorer_return=47.01 vision=(0.000,0.000) new_fields=(82.4,65.1) distance=6.20 loss=(-0.001,3.659) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=2.25 chaser_return=6.50 explorer_return=49.38 vision=(0.000,0.000) new_fields=(83.2,66.8) distance=5.63 loss=(-0.001,3.732) kl=0.0002
+update=6500 collisions=1.83 chaser_return=6.26 explorer_return=45.66 vision=(0.000,0.000) new_fields=(81.9,64.3) distance=6.03 loss=(-0.000,3.905) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=2.28 chaser_return=6.52 explorer_return=49.94 vision=(0.000,0.000) new_fields=(83.4,67.1) distance=5.44 loss=(-0.001,3.666) kl=0.0006
+update=7500 collisions=1.73 chaser_return=6.92 explorer_return=58.75 vision=(0.000,0.000) new_fields=(85.2,72.9) distance=5.51 loss=(-0.001,3.484) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=2.67 chaser_return=6.61 explorer_return=56.50 vision=(0.000,0.000) new_fields=(83.9,71.6) distance=5.52 loss=(-0.000,3.613) kl=0.0003
+update=8500 collisions=1.50 chaser_return=6.84 explorer_return=55.45 vision=(0.000,0.000) new_fields=(84.6,70.6) distance=5.27 loss=(-0.001,3.571) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=1.60 chaser_return=6.97 explorer_return=57.17 vision=(0.000,0.000) new_fields=(85.3,71.8) distance=6.14 loss=(-0.001,3.552) kl=0.0001
+update=9500 collisions=1.62 chaser_return=7.01 explorer_return=58.23 vision=(0.000,0.000) new_fields=(85.6,72.5) distance=5.88 loss=(-0.001,3.518) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=1.65 chaser_return=6.77 explorer_return=54.14 vision=(0.000,0.000) new_fields=(84.4,69.8) distance=5.77 loss=(-0.001,3.711) kl=0.0004
+update=10500 collisions=2.00 chaser_return=6.75 explorer_return=54.81 vision=(0.000,0.000) new_fields=(84.3,70.3) distance=5.62 loss=(-0.001,3.519) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=1.50 chaser_return=6.85 explorer_return=57.17 vision=(0.000,0.000) new_fields=(84.8,71.9) distance=5.25 loss=(0.000,3.642) kl=-0.0000
+update=11500 collisions=1.38 chaser_return=7.05 explorer_return=54.33 vision=(0.000,0.000) new_fields=(85.6,70.0) distance=5.77 loss=(-0.001,3.390) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=1.77 chaser_return=7.19 explorer_return=54.14 vision=(0.000,0.000) new_fields=(86.6,69.8) distance=6.10 loss=(-0.001,3.461) kl=0.0003
+update=12500 collisions=1.50 chaser_return=7.08 explorer_return=53.39 vision=(0.000,0.000) new_fields=(85.8,69.4) distance=5.96 loss=(-0.001,3.413) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=1.68 chaser_return=7.19 explorer_return=53.20 vision=(0.000,0.000) new_fields=(86.6,69.3) distance=6.22 loss=(-0.001,3.548) kl=0.0001
+update=13500 collisions=2.00 chaser_return=7.20 explorer_return=50.84 vision=(0.000,0.000) new_fields=(86.7,67.7) distance=5.75 loss=(-0.000,3.510) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=3.50 chaser_return=6.98 explorer_return=50.20 vision=(0.000,0.000) new_fields=(85.6,67.4) distance=4.92 loss=(-0.001,3.714) kl=0.0000
+update=14500 collisions=2.12 chaser_return=6.22 explorer_return=46.08 vision=(0.000,0.000) new_fields=(81.8,64.5) distance=3.76 loss=(-0.001,3.703) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=1.88 chaser_return=6.36 explorer_return=48.62 vision=(0.000,0.000) new_fields=(82.3,66.1) distance=4.35 loss=(-0.001,3.744) kl=0.0000
+update=15500 collisions=1.98 chaser_return=6.39 explorer_return=40.26 vision=(0.000,0.000) new_fields=(82.6,60.7) distance=4.36 loss=(-0.001,3.929) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=1.90 chaser_return=6.93 explorer_return=49.94 vision=(0.000,0.000) new_fields=(85.2,67.0) distance=4.74 loss=(-0.001,3.416) kl=0.0002
+update=16500 collisions=1.77 chaser_return=6.88 explorer_return=49.90 vision=(0.000,0.000) new_fields=(84.9,67.0) distance=4.16 loss=(-0.001,3.580) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=1.62 chaser_return=7.02 explorer_return=61.79 vision=(0.000,0.000) new_fields=(85.6,75.1) distance=4.57 loss=(-0.001,3.388) kl=0.0002
+update=17500 collisions=2.05 chaser_return=7.17 explorer_return=51.29 vision=(0.000,0.000) new_fields=(86.5,68.1) distance=4.23 loss=(-0.001,3.639) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=2.67 chaser_return=6.88 explorer_return=54.51 vision=(0.000,0.000) new_fields=(85.2,70.3) distance=4.54 loss=(-0.001,3.795) kl=0.0003
+update=18500 collisions=2.00 chaser_return=7.42 explorer_return=58.86 vision=(0.000,0.000) new_fields=(87.7,73.1) distance=3.79 loss=(-0.001,3.571) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=2.15 chaser_return=7.32 explorer_return=60.21 vision=(0.000,0.000) new_fields=(87.3,74.0) distance=4.64 loss=(-0.000,3.456) kl=0.0005
+update=19500 collisions=1.90 chaser_return=7.38 explorer_return=62.76 vision=(0.000,0.000) new_fields=(87.6,75.7) distance=3.78 loss=(-0.001,3.499) kl=0.0006
+update=20000 collisions=1.73 chaser_return=7.30 explorer_return=62.12 vision=(0.000,0.000) new_fields=(87.1,75.2) distance=4.14 loss=(-0.001,3.491) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7631.695038600941, "metrics": {"approx_kl": 0.0002841827333717878, "chaser_grad_norm": 0.04017048981040716, "chaser_new_fields": 87.07500457763672, "chaser_partner_vision": 0.0, "chaser_return": 7.300001621246338, "chaser_subspace_norm": 112.96394348144531, "collisions_per_episode": 1.725000023841858, "entropy": 0.024153232108801603, "explorer_grad_norm": 0.8466866165399551, "explorer_new_fields": 75.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 62.125, "explorer_subspace_norm": 36.111541748046875, "final_distance": 4.144174098968506, "max_param_abs": 1.6294050216674805, "policy_loss": -0.0005663236785924397, "value_loss": 3.4905534386634827}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01", "seed": 3, "task": "non_social", "unit_id": "non_social/seed_0003"}
+=== exited 2026-07-02T09:43:38.602587+00:00 returncode=0 ===

logs/non_social_seed_0004_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.525938+00:00 task=non_social seed=4 attempt=1 ===
+update=0001 collisions=1.52 chaser_return=-2.70 explorer_return=2.12 vision=(0.000,0.000) new_fields=(36.8,35.0) distance=5.16 loss=(-0.001,3.118) kl=-0.0000
+update=0500 collisions=2.80 chaser_return=4.54 explorer_return=44.31 vision=(0.000,0.000) new_fields=(73.3,63.2) distance=5.77 loss=(-0.002,4.773) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=3.12 chaser_return=4.64 explorer_return=46.60 vision=(0.000,0.000) new_fields=(73.9,65.0) distance=5.48 loss=(-0.001,4.439) kl=0.0003
+update=1500 collisions=2.67 chaser_return=5.74 explorer_return=51.51 vision=(0.000,0.000) new_fields=(79.4,68.2) distance=5.40 loss=(-0.001,4.171) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=2.62 chaser_return=5.74 explorer_return=47.46 vision=(0.000,0.000) new_fields=(79.3,65.6) distance=4.74 loss=(-0.001,4.247) kl=0.0001
+update=2500 collisions=2.03 chaser_return=5.46 explorer_return=46.90 vision=(0.000,0.000) new_fields=(77.7,65.3) distance=5.53 loss=(-0.001,4.049) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=3.23 chaser_return=5.51 explorer_return=47.58 vision=(0.000,0.000) new_fields=(78.2,65.7) distance=5.26 loss=(-0.002,4.019) kl=0.0002
+update=3500 collisions=1.75 chaser_return=6.24 explorer_return=51.29 vision=(0.000,0.000) new_fields=(81.8,68.0) distance=5.07 loss=(-0.001,3.856) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=1.80 chaser_return=5.94 explorer_return=53.16 vision=(0.000,0.000) new_fields=(80.2,69.2) distance=4.02 loss=(-0.000,3.730) kl=0.0000
+update=4500 collisions=1.88 chaser_return=6.07 explorer_return=54.40 vision=(0.000,0.000) new_fields=(80.8,70.1) distance=5.64 loss=(-0.001,3.669) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=2.05 chaser_return=6.00 explorer_return=56.16 vision=(0.000,0.000) new_fields=(80.7,71.1) distance=4.47 loss=(-0.001,3.657) kl=0.0005
+update=5500 collisions=2.33 chaser_return=6.25 explorer_return=57.17 vision=(0.000,0.000) new_fields=(81.8,72.0) distance=4.90 loss=(-0.001,3.611) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=1.80 chaser_return=6.38 explorer_return=57.36 vision=(0.000,0.000) new_fields=(82.5,72.0) distance=4.56 loss=(-0.001,3.647) kl=0.0004
+update=6500 collisions=2.48 chaser_return=5.81 explorer_return=57.70 vision=(0.000,0.000) new_fields=(79.7,72.3) distance=5.06 loss=(-0.001,3.639) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=1.93 chaser_return=6.27 explorer_return=57.74 vision=(0.000,0.000) new_fields=(82.0,72.4) distance=4.56 loss=(-0.000,3.609) kl=-0.0001
+update=7500 collisions=2.00 chaser_return=5.67 explorer_return=56.01 vision=(0.000,0.000) new_fields=(78.9,71.2) distance=5.03 loss=(-0.001,3.669) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=2.30 chaser_return=6.02 explorer_return=55.38 vision=(0.000,0.000) new_fields=(80.9,70.7) distance=5.81 loss=(-0.001,3.602) kl=0.0003
+update=8500 collisions=4.60 chaser_return=4.99 explorer_return=50.16 vision=(0.000,0.000) new_fields=(75.7,67.3) distance=4.70 loss=(-0.001,3.843) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=2.70 chaser_return=5.79 explorer_return=54.81 vision=(0.000,0.000) new_fields=(79.7,70.5) distance=4.68 loss=(-0.000,3.554) kl=0.0002
+update=9500 collisions=2.12 chaser_return=5.53 explorer_return=53.35 vision=(0.000,0.000) new_fields=(78.2,69.5) distance=5.71 loss=(-0.001,3.767) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=2.30 chaser_return=5.39 explorer_return=58.38 vision=(0.000,0.000) new_fields=(77.6,72.9) distance=5.03 loss=(-0.001,3.573) kl=0.0008
+update=10500 collisions=1.27 chaser_return=5.81 explorer_return=58.53 vision=(0.000,0.000) new_fields=(79.4,72.8) distance=5.08 loss=(-0.001,3.605) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=2.23 chaser_return=5.72 explorer_return=59.54 vision=(0.000,0.000) new_fields=(79.3,73.5) distance=5.25 loss=(-0.001,3.539) kl=-0.0002
+update=11500 collisions=1.88 chaser_return=5.56 explorer_return=57.03 vision=(0.000,0.000) new_fields=(78.4,71.9) distance=5.20 loss=(-0.001,3.630) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=2.08 chaser_return=5.09 explorer_return=55.75 vision=(0.000,0.000) new_fields=(75.9,71.1) distance=5.03 loss=(-0.001,3.760) kl=0.0006
+update=12500 collisions=1.95 chaser_return=6.06 explorer_return=56.46 vision=(0.000,0.000) new_fields=(80.9,71.3) distance=5.68 loss=(-0.001,3.521) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=2.10 chaser_return=5.33 explorer_return=54.92 vision=(0.000,0.000) new_fields=(77.2,70.5) distance=5.22 loss=(-0.001,3.856) kl=0.0009
+update=13500 collisions=2.23 chaser_return=5.47 explorer_return=59.76 vision=(0.000,0.000) new_fields=(78.0,73.6) distance=4.86 loss=(-0.001,3.574) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=1.90 chaser_return=5.82 explorer_return=60.44 vision=(0.000,0.000) new_fields=(79.5,74.0) distance=4.30 loss=(-0.001,3.546) kl=0.0004
+update=14500 collisions=2.28 chaser_return=5.39 explorer_return=59.42 vision=(0.000,0.000) new_fields=(77.4,73.5) distance=5.56 loss=(-0.000,3.604) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=1.85 chaser_return=6.00 explorer_return=58.19 vision=(0.000,0.000) new_fields=(80.5,72.6) distance=4.86 loss=(-0.001,3.492) kl=0.0001
+update=15500 collisions=1.93 chaser_return=6.20 explorer_return=57.81 vision=(0.000,0.000) new_fields=(81.8,72.4) distance=5.56 loss=(-0.001,3.532) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=2.00 chaser_return=6.21 explorer_return=57.51 vision=(0.000,0.000) new_fields=(81.6,72.1) distance=4.76 loss=(-0.001,3.554) kl=0.0004
+update=16500 collisions=2.03 chaser_return=5.92 explorer_return=59.84 vision=(0.000,0.000) new_fields=(80.4,73.8) distance=4.73 loss=(-0.001,3.486) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=1.60 chaser_return=5.77 explorer_return=57.81 vision=(0.000,0.000) new_fields=(79.4,72.3) distance=5.04 loss=(-0.001,3.596) kl=0.0001
+update=17500 collisions=1.55 chaser_return=6.05 explorer_return=61.90 vision=(0.000,0.000) new_fields=(80.6,75.1) distance=5.21 loss=(-0.001,3.360) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=2.35 chaser_return=5.90 explorer_return=63.17 vision=(0.000,0.000) new_fields=(80.1,76.1) distance=5.27 loss=(-0.001,3.491) kl=0.0002
+update=18500 collisions=1.60 chaser_return=6.21 explorer_return=63.40 vision=(0.000,0.000) new_fields=(81.5,76.1) distance=5.32 loss=(-0.001,3.353) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=1.68 chaser_return=5.94 explorer_return=61.30 vision=(0.000,0.000) new_fields=(80.2,74.7) distance=5.57 loss=(-0.001,3.479) kl=0.0003
+update=19500 collisions=2.23 chaser_return=5.84 explorer_return=62.58 vision=(0.000,0.000) new_fields=(79.8,75.6) distance=5.48 loss=(-0.000,3.403) kl=0.0004
+update=20000 collisions=1.83 chaser_return=6.33 explorer_return=66.10 vision=(0.000,0.000) new_fields=(82.2,78.0) distance=5.31 loss=(-0.001,3.323) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7640.99740531703, "metrics": {"approx_kl": 0.00037036755928841064, "chaser_grad_norm": 0.08058446925133467, "chaser_new_fields": 82.17500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.330001354217529, "chaser_subspace_norm": 92.44800567626953, "collisions_per_episode": 1.8250000476837158, "entropy": 0.04575740918517113, "explorer_grad_norm": 0.3279200941324234, "explorer_new_fields": 77.9749984741211, "explorer_partner_vision": 0.0, "explorer_return": 66.0999984741211, "explorer_subspace_norm": 29.62271499633789, "final_distance": 5.308696269989014, "max_param_abs": 1.3726928234100342, "policy_loss": -0.0006299224167634065, "value_loss": 3.322889506816864}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01", "seed": 4, "task": "non_social", "unit_id": "non_social/seed_0004"}
+=== exited 2026-07-02T09:43:48.667830+00:00 returncode=0 ===

logs/non_social_seed_0005_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:28.527565+00:00 task=non_social seed=5 attempt=1 ===
+update=0001 collisions=1.38 chaser_return=-3.43 explorer_return=1.86 vision=(0.000,0.000) new_fields=(33.1,34.8) distance=5.59 loss=(-0.001,3.093) kl=0.0000
+update=0500 collisions=1.48 chaser_return=5.37 explorer_return=54.59 vision=(0.000,0.000) new_fields=(77.4,70.1) distance=4.62 loss=(-0.002,4.404) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=1.60 chaser_return=5.86 explorer_return=56.39 vision=(0.000,0.000) new_fields=(79.8,71.3) distance=5.76 loss=(-0.001,4.144) kl=0.0001
+update=1500 collisions=0.73 chaser_return=6.31 explorer_return=56.24 vision=(0.000,0.000) new_fields=(81.8,71.0) distance=5.41 loss=(-0.001,3.992) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=1.35 chaser_return=5.83 explorer_return=56.01 vision=(0.000,0.000) new_fields=(79.5,71.0) distance=4.97 loss=(-0.001,4.067) kl=0.0005
+update=2500 collisions=0.82 chaser_return=6.57 explorer_return=60.06 vision=(0.000,0.000) new_fields=(83.0,73.6) distance=4.89 loss=(-0.002,4.075) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=1.18 chaser_return=6.21 explorer_return=59.69 vision=(0.000,0.000) new_fields=(81.2,73.5) distance=4.73 loss=(-0.001,4.040) kl=-0.0001
+update=3500 collisions=1.33 chaser_return=6.47 explorer_return=61.23 vision=(0.000,0.000) new_fields=(82.8,74.6) distance=4.77 loss=(-0.001,3.945) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=1.75 chaser_return=6.95 explorer_return=61.23 vision=(0.000,0.000) new_fields=(85.3,74.7) distance=4.62 loss=(-0.000,3.940) kl=-0.0001
+update=4500 collisions=1.43 chaser_return=6.96 explorer_return=60.89 vision=(0.000,0.000) new_fields=(85.2,74.2) distance=5.47 loss=(-0.001,3.911) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=1.20 chaser_return=6.60 explorer_return=61.34 vision=(0.000,0.000) new_fields=(83.3,74.5) distance=4.56 loss=(-0.001,3.877) kl=0.0002
+update=5500 collisions=0.93 chaser_return=5.47 explorer_return=61.79 vision=(0.000,0.000) new_fields=(77.6,74.8) distance=5.23 loss=(-0.001,3.934) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=1.02 chaser_return=5.31 explorer_return=60.21 vision=(0.000,0.000) new_fields=(77.0,73.7) distance=5.47 loss=(-0.000,3.929) kl=0.0001
+update=6500 collisions=1.18 chaser_return=4.91 explorer_return=59.05 vision=(0.000,0.000) new_fields=(75.0,73.0) distance=4.71 loss=(-0.001,3.863) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=1.38 chaser_return=5.28 explorer_return=63.17 vision=(0.000,0.000) new_fields=(76.8,75.8) distance=5.45 loss=(-0.000,3.864) kl=0.0001
+update=7500 collisions=1.43 chaser_return=5.53 explorer_return=64.90 vision=(0.000,0.000) new_fields=(78.0,77.0) distance=4.91 loss=(-0.001,3.865) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=1.68 chaser_return=5.52 explorer_return=61.45 vision=(0.000,0.000) new_fields=(78.0,74.7) distance=4.85 loss=(-0.001,3.723) kl=0.0004
+update=8500 collisions=1.08 chaser_return=5.36 explorer_return=65.43 vision=(0.000,0.000) new_fields=(77.0,77.3) distance=4.80 loss=(-0.001,3.755) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=1.10 chaser_return=5.75 explorer_return=63.70 vision=(0.000,0.000) new_fields=(78.9,76.2) distance=5.00 loss=(-0.001,3.859) kl=0.0004
+update=9500 collisions=0.98 chaser_return=6.59 explorer_return=61.19 vision=(0.000,0.000) new_fields=(83.2,74.5) distance=4.30 loss=(-0.001,3.872) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=1.05 chaser_return=6.06 explorer_return=64.00 vision=(0.000,0.000) new_fields=(80.5,76.3) distance=5.21 loss=(-0.001,3.785) kl=0.0003
+update=10500 collisions=0.90 chaser_return=5.39 explorer_return=63.17 vision=(0.000,0.000) new_fields=(77.2,75.7) distance=5.27 loss=(-0.000,3.767) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=1.18 chaser_return=5.79 explorer_return=61.23 vision=(0.000,0.000) new_fields=(79.2,74.4) distance=5.14 loss=(-0.000,3.715) kl=0.0001
+update=11500 collisions=1.30 chaser_return=6.08 explorer_return=61.23 vision=(0.000,0.000) new_fields=(80.8,74.6) distance=5.31 loss=(-0.001,3.808) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=1.20 chaser_return=5.33 explorer_return=64.83 vision=(0.000,0.000) new_fields=(77.0,76.8) distance=4.53 loss=(-0.001,3.822) kl=0.0001
+update=12500 collisions=1.33 chaser_return=5.14 explorer_return=63.44 vision=(0.000,0.000) new_fields=(76.0,76.0) distance=5.12 loss=(-0.000,3.905) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=0.82 chaser_return=5.30 explorer_return=61.34 vision=(0.000,0.000) new_fields=(76.8,74.5) distance=4.83 loss=(-0.001,3.705) kl=-0.0002
+update=13500 collisions=0.70 chaser_return=4.78 explorer_return=64.71 vision=(0.000,0.000) new_fields=(74.2,76.7) distance=4.84 loss=(-0.001,3.741) kl=-0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=0.85 chaser_return=4.31 explorer_return=60.48 vision=(0.000,0.000) new_fields=(71.8,74.0) distance=4.77 loss=(-0.001,3.648) kl=0.0004
+update=14500 collisions=0.95 chaser_return=4.47 explorer_return=61.53 vision=(0.000,0.000) new_fields=(72.7,74.6) distance=4.75 loss=(-0.000,3.682) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=1.43 chaser_return=4.67 explorer_return=62.54 vision=(0.000,0.000) new_fields=(73.7,75.4) distance=4.49 loss=(-0.001,3.678) kl=0.0005
+update=15500 collisions=1.25 chaser_return=4.96 explorer_return=59.58 vision=(0.000,0.000) new_fields=(75.2,73.5) distance=4.57 loss=(-0.001,3.831) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=1.10 chaser_return=4.91 explorer_return=61.56 vision=(0.000,0.000) new_fields=(74.8,74.8) distance=3.97 loss=(-0.000,3.616) kl=0.0004
+update=16500 collisions=1.43 chaser_return=4.88 explorer_return=59.31 vision=(0.000,0.000) new_fields=(74.9,73.4) distance=4.76 loss=(-0.000,3.797) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=1.62 chaser_return=5.41 explorer_return=58.67 vision=(0.000,0.000) new_fields=(77.6,73.0) distance=5.04 loss=(-0.001,3.581) kl=0.0003
+update=17500 collisions=1.12 chaser_return=4.96 explorer_return=62.99 vision=(0.000,0.000) new_fields=(75.1,75.8) distance=5.48 loss=(-0.001,3.771) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=1.23 chaser_return=5.24 explorer_return=61.41 vision=(0.000,0.000) new_fields=(76.6,74.7) distance=4.87 loss=(-0.001,3.698) kl=0.0004
+update=18500 collisions=1.52 chaser_return=5.74 explorer_return=61.08 vision=(0.000,0.000) new_fields=(79.2,74.5) distance=5.09 loss=(-0.001,3.678) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=1.33 chaser_return=5.32 explorer_return=62.61 vision=(0.000,0.000) new_fields=(76.9,75.5) distance=4.53 loss=(-0.000,3.703) kl=0.0003
+update=19500 collisions=0.73 chaser_return=5.46 explorer_return=63.33 vision=(0.000,0.000) new_fields=(77.5,75.7) distance=4.44 loss=(-0.001,3.721) kl=0.0006
+update=20000 collisions=1.02 chaser_return=6.02 explorer_return=59.42 vision=(0.000,0.000) new_fields=(80.5,73.2) distance=4.73 loss=(-0.000,3.523) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7562.32880273601, "metrics": {"approx_kl": 0.00022429173202174373, "chaser_grad_norm": 0.04249562695622444, "chaser_new_fields": 80.4749984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.015000820159912, "chaser_subspace_norm": 93.73216247558594, "collisions_per_episode": 1.024999976158142, "entropy": 0.05529557354748249, "explorer_grad_norm": 0.39152009785175323, "explorer_new_fields": 73.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 59.42499923706055, "explorer_subspace_norm": 117.07512664794922, "final_distance": 4.730400562286377, "max_param_abs": 1.606895923614502, "policy_loss": -0.0002411644566997495, "value_loss": 3.523293375968933}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01", "seed": 5, "task": "non_social", "unit_id": "non_social/seed_0005"}
+=== exited 2026-07-02T11:49:40.894281+00:00 returncode=0 ===

logs/non_social_seed_0006_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:48.617227+00:00 task=non_social seed=6 attempt=1 ===
+update=0001 collisions=2.35 chaser_return=-3.20 explorer_return=-0.43 vision=(0.000,0.000) new_fields=(34.4,33.5) distance=4.61 loss=(-0.001,3.030) kl=0.0000
+update=0500 collisions=0.75 chaser_return=5.09 explorer_return=43.26 vision=(0.000,0.000) new_fields=(75.6,62.5) distance=5.97 loss=(-0.002,4.700) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=0.95 chaser_return=5.83 explorer_return=40.90 vision=(0.000,0.000) new_fields=(79.4,60.9) distance=5.96 loss=(-0.001,4.234) kl=0.0002
+update=1500 collisions=0.90 chaser_return=6.05 explorer_return=38.84 vision=(0.000,0.000) new_fields=(80.5,59.5) distance=5.93 loss=(-0.001,4.259) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=0.60 chaser_return=5.84 explorer_return=39.36 vision=(0.000,0.000) new_fields=(79.3,59.7) distance=5.52 loss=(-0.001,4.038) kl=0.0002
+update=2500 collisions=0.53 chaser_return=6.87 explorer_return=41.58 vision=(0.000,0.000) new_fields=(84.5,61.2) distance=5.54 loss=(-0.000,3.781) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=0.62 chaser_return=6.37 explorer_return=43.83 vision=(0.000,0.000) new_fields=(82.1,62.7) distance=5.62 loss=(-0.001,4.088) kl=-0.0001
+update=3500 collisions=0.65 chaser_return=6.87 explorer_return=38.95 vision=(0.000,0.000) new_fields=(84.6,59.4) distance=5.97 loss=(-0.000,3.927) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=0.88 chaser_return=6.80 explorer_return=43.75 vision=(0.000,0.000) new_fields=(84.2,62.7) distance=5.37 loss=(-0.001,3.941) kl=-0.0000
+update=4500 collisions=0.62 chaser_return=6.63 explorer_return=41.61 vision=(0.000,0.000) new_fields=(83.4,61.2) distance=5.82 loss=(-0.001,3.979) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=1.43 chaser_return=6.47 explorer_return=38.35 vision=(0.000,0.000) new_fields=(82.8,59.4) distance=4.96 loss=(-0.001,3.918) kl=0.0001
+update=5500 collisions=1.10 chaser_return=6.12 explorer_return=42.81 vision=(0.000,0.000) new_fields=(81.0,62.1) distance=4.76 loss=(-0.001,3.662) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=1.23 chaser_return=6.65 explorer_return=41.76 vision=(0.000,0.000) new_fields=(83.6,61.6) distance=5.43 loss=(-0.001,3.932) kl=0.0006
+update=6500 collisions=0.78 chaser_return=6.62 explorer_return=45.03 vision=(0.000,0.000) new_fields=(83.5,63.6) distance=5.08 loss=(-0.000,3.829) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=1.27 chaser_return=6.36 explorer_return=45.10 vision=(0.000,0.000) new_fields=(82.3,63.7) distance=5.44 loss=(-0.001,3.815) kl=0.0003
+update=7500 collisions=1.05 chaser_return=5.68 explorer_return=46.53 vision=(0.000,0.000) new_fields=(78.7,64.6) distance=5.28 loss=(-0.001,3.874) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=0.70 chaser_return=5.67 explorer_return=46.19 vision=(0.000,0.000) new_fields=(78.5,64.4) distance=4.73 loss=(-0.001,3.969) kl=0.0007
+update=8500 collisions=0.98 chaser_return=6.50 explorer_return=53.05 vision=(0.000,0.000) new_fields=(83.0,68.9) distance=4.71 loss=(-0.001,3.682) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=1.10 chaser_return=5.61 explorer_return=55.11 vision=(0.000,0.000) new_fields=(78.4,70.5) distance=4.92 loss=(-0.001,3.816) kl=0.0003
+update=9500 collisions=0.70 chaser_return=6.09 explorer_return=48.96 vision=(0.000,0.000) new_fields=(80.6,66.2) distance=5.17 loss=(-0.001,3.700) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=0.70 chaser_return=5.81 explorer_return=54.25 vision=(0.000,0.000) new_fields=(79.3,69.6) distance=4.84 loss=(0.000,3.720) kl=0.0002
+update=10500 collisions=1.08 chaser_return=6.25 explorer_return=52.86 vision=(0.000,0.000) new_fields=(81.6,69.0) distance=4.62 loss=(-0.001,3.482) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=0.55 chaser_return=6.15 explorer_return=49.90 vision=(0.000,0.000) new_fields=(80.8,66.7) distance=4.71 loss=(-0.001,3.714) kl=0.0001
+update=11500 collisions=1.12 chaser_return=6.43 explorer_return=38.09 vision=(0.000,0.000) new_fields=(82.5,58.9) distance=4.79 loss=(-0.001,4.131) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=0.82 chaser_return=6.47 explorer_return=37.45 vision=(0.000,0.000) new_fields=(82.6,58.4) distance=4.44 loss=(-0.001,3.982) kl=0.0005
+update=12500 collisions=0.85 chaser_return=6.77 explorer_return=42.03 vision=(0.000,0.000) new_fields=(84.2,61.5) distance=4.87 loss=(-0.001,3.788) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=1.05 chaser_return=5.94 explorer_return=44.54 vision=(0.000,0.000) new_fields=(80.1,63.2) distance=5.31 loss=(-0.001,4.082) kl=0.0004
+update=13500 collisions=1.35 chaser_return=6.58 explorer_return=44.24 vision=(0.000,0.000) new_fields=(83.4,63.2) distance=4.13 loss=(-0.000,3.821) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=1.98 chaser_return=6.18 explorer_return=40.94 vision=(0.000,0.000) new_fields=(81.5,61.1) distance=4.75 loss=(-0.001,4.174) kl=0.0006
+update=14500 collisions=1.55 chaser_return=6.05 explorer_return=40.45 vision=(0.000,0.000) new_fields=(80.6,60.6) distance=4.70 loss=(-0.000,3.985) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=1.15 chaser_return=6.25 explorer_return=38.69 vision=(0.000,0.000) new_fields=(81.5,59.5) distance=4.01 loss=(-0.000,3.674) kl=0.0004
+update=15500 collisions=1.27 chaser_return=6.11 explorer_return=41.31 vision=(0.000,0.000) new_fields=(80.9,61.2) distance=3.81 loss=(-0.001,3.731) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=1.93 chaser_return=5.63 explorer_return=43.41 vision=(0.000,0.000) new_fields=(78.7,62.7) distance=4.86 loss=(-0.001,3.888) kl=0.0003
+update=16500 collisions=1.20 chaser_return=6.00 explorer_return=40.64 vision=(0.000,0.000) new_fields=(80.2,60.7) distance=4.30 loss=(-0.000,3.691) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=1.12 chaser_return=5.92 explorer_return=40.75 vision=(0.000,0.000) new_fields=(79.9,60.9) distance=4.73 loss=(-0.001,3.881) kl=0.0006
+update=17500 collisions=0.95 chaser_return=6.14 explorer_return=41.35 vision=(0.000,0.000) new_fields=(81.0,61.3) distance=4.91 loss=(-0.000,3.892) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=0.93 chaser_return=6.10 explorer_return=42.92 vision=(0.000,0.000) new_fields=(80.8,62.2) distance=4.83 loss=(-0.001,3.887) kl=0.0004
+update=18500 collisions=1.25 chaser_return=6.06 explorer_return=42.96 vision=(0.000,0.000) new_fields=(80.7,62.3) distance=4.71 loss=(-0.000,3.709) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=1.58 chaser_return=6.40 explorer_return=41.65 vision=(0.000,0.000) new_fields=(82.4,61.5) distance=4.66 loss=(-0.001,3.625) kl=0.0004
+update=19500 collisions=1.65 chaser_return=6.14 explorer_return=42.66 vision=(0.000,0.000) new_fields=(81.4,62.2) distance=4.72 loss=(-0.001,3.940) kl=0.0006
+update=20000 collisions=1.15 chaser_return=6.39 explorer_return=41.24 vision=(0.000,0.000) new_fields=(82.4,61.0) distance=4.67 loss=(-0.000,3.841) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7567.213006881066, "metrics": {"approx_kl": 8.635803430578903e-06, "chaser_grad_norm": 0.10996681079268456, "chaser_new_fields": 82.375, "chaser_partner_vision": 0.0, "chaser_return": 6.390000820159912, "chaser_subspace_norm": 73.99776458740234, "collisions_per_episode": 1.149999976158142, "entropy": 0.05159799661487341, "explorer_grad_norm": 0.7565127648413181, "explorer_new_fields": 61.025001525878906, "explorer_partner_vision": 0.0, "explorer_return": 41.23749923706055, "explorer_subspace_norm": 46.546077728271484, "final_distance": 4.668344020843506, "max_param_abs": 1.2401388883590698, "policy_loss": -0.0002673804239634592, "value_loss": 3.8410890698432922}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01", "seed": 6, "task": "non_social", "unit_id": "non_social/seed_0006"}
+=== exited 2026-07-02T11:50:00.915352+00:00 returncode=0 ===

logs/non_social_seed_0007_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:58.681826+00:00 task=non_social seed=7 attempt=1 ===
+update=0001 collisions=1.48 chaser_return=-3.24 explorer_return=1.49 vision=(0.000,0.000) new_fields=(34.1,34.5) distance=5.13 loss=(-0.000,3.129) kl=0.0000
+update=0500 collisions=3.03 chaser_return=4.67 explorer_return=45.03 vision=(0.000,0.000) new_fields=(74.1,64.0) distance=5.01 loss=(-0.002,4.623) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=2.88 chaser_return=4.99 explorer_return=46.23 vision=(0.000,0.000) new_fields=(75.7,64.7) distance=5.35 loss=(-0.001,4.559) kl=0.0003
+update=1500 collisions=2.53 chaser_return=5.85 explorer_return=51.25 vision=(0.000,0.000) new_fields=(79.9,68.1) distance=4.94 loss=(-0.001,4.427) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=2.88 chaser_return=5.68 explorer_return=56.42 vision=(0.000,0.000) new_fields=(79.4,71.6) distance=6.11 loss=(-0.001,4.207) kl=0.0001
+update=2500 collisions=1.88 chaser_return=5.87 explorer_return=61.38 vision=(0.000,0.000) new_fields=(79.8,74.8) distance=4.83 loss=(-0.001,3.954) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=2.40 chaser_return=4.89 explorer_return=62.54 vision=(0.000,0.000) new_fields=(75.1,75.5) distance=4.51 loss=(-0.000,4.105) kl=0.0001
+update=3500 collisions=2.42 chaser_return=6.16 explorer_return=59.39 vision=(0.000,0.000) new_fields=(81.4,73.4) distance=5.17 loss=(-0.001,3.993) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=2.55 chaser_return=5.89 explorer_return=65.65 vision=(0.000,0.000) new_fields=(80.1,77.8) distance=5.10 loss=(-0.001,3.974) kl=0.0002
+update=4500 collisions=2.55 chaser_return=5.20 explorer_return=61.83 vision=(0.000,0.000) new_fields=(76.5,75.1) distance=5.03 loss=(-0.000,4.066) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=2.28 chaser_return=4.79 explorer_return=64.53 vision=(0.000,0.000) new_fields=(74.7,76.8) distance=5.29 loss=(-0.001,3.989) kl=0.0005
+update=5500 collisions=2.70 chaser_return=5.46 explorer_return=68.16 vision=(0.000,0.000) new_fields=(78.1,79.5) distance=4.66 loss=(-0.001,3.927) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=2.25 chaser_return=5.89 explorer_return=65.91 vision=(0.000,0.000) new_fields=(80.2,77.7) distance=4.68 loss=(-0.001,3.832) kl=0.0004
+update=6500 collisions=2.33 chaser_return=5.18 explorer_return=63.92 vision=(0.000,0.000) new_fields=(76.4,76.3) distance=4.94 loss=(-0.001,4.087) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=2.17 chaser_return=5.18 explorer_return=66.21 vision=(0.000,0.000) new_fields=(76.4,78.0) distance=4.88 loss=(-0.001,4.070) kl=0.0006
+update=7500 collisions=4.03 chaser_return=5.55 explorer_return=62.01 vision=(0.000,0.000) new_fields=(78.5,75.2) distance=5.08 loss=(-0.001,3.999) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=2.62 chaser_return=5.84 explorer_return=68.09 vision=(0.000,0.000) new_fields=(80.1,79.1) distance=4.90 loss=(-0.001,3.641) kl=0.0002
+update=8500 collisions=2.60 chaser_return=6.58 explorer_return=63.62 vision=(0.000,0.000) new_fields=(83.7,76.2) distance=5.30 loss=(-0.001,3.789) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=2.05 chaser_return=5.99 explorer_return=60.66 vision=(0.000,0.000) new_fields=(80.6,74.3) distance=4.89 loss=(-0.001,3.983) kl=0.0002
+update=9500 collisions=2.48 chaser_return=6.74 explorer_return=62.46 vision=(0.000,0.000) new_fields=(84.5,75.3) distance=4.45 loss=(-0.001,3.623) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=3.20 chaser_return=5.66 explorer_return=61.19 vision=(0.000,0.000) new_fields=(79.1,74.7) distance=4.18 loss=(-0.001,3.959) kl=0.0001
+update=10500 collisions=2.75 chaser_return=6.05 explorer_return=61.23 vision=(0.000,0.000) new_fields=(81.0,74.7) distance=4.20 loss=(-0.000,3.825) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=2.48 chaser_return=6.60 explorer_return=58.56 vision=(0.000,0.000) new_fields=(83.6,72.8) distance=4.32 loss=(-0.000,3.884) kl=0.0002
+update=11500 collisions=2.62 chaser_return=6.16 explorer_return=66.81 vision=(0.000,0.000) new_fields=(81.5,78.3) distance=4.88 loss=(-0.000,3.705) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=2.62 chaser_return=6.47 explorer_return=67.75 vision=(0.000,0.000) new_fields=(83.0,79.0) distance=5.42 loss=(-0.001,3.711) kl=0.0005
+update=12500 collisions=2.23 chaser_return=5.84 explorer_return=61.90 vision=(0.000,0.000) new_fields=(79.7,75.0) distance=5.01 loss=(-0.001,3.898) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=2.73 chaser_return=5.98 explorer_return=63.74 vision=(0.000,0.000) new_fields=(80.5,76.3) distance=5.38 loss=(-0.001,3.841) kl=0.0003
+update=13500 collisions=2.05 chaser_return=6.51 explorer_return=71.61 vision=(0.000,0.000) new_fields=(83.3,81.6) distance=4.74 loss=(-0.000,3.638) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=2.67 chaser_return=6.09 explorer_return=74.01 vision=(0.000,0.000) new_fields=(81.1,83.3) distance=4.75 loss=(-0.000,3.677) kl=0.0001
+update=14500 collisions=3.23 chaser_return=5.49 explorer_return=67.41 vision=(0.000,0.000) new_fields=(78.4,79.1) distance=4.86 loss=(-0.001,3.808) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=2.67 chaser_return=5.91 explorer_return=73.30 vision=(0.000,0.000) new_fields=(80.3,83.0) distance=5.65 loss=(-0.001,3.792) kl=0.0002
+update=15500 collisions=2.12 chaser_return=5.64 explorer_return=70.15 vision=(0.000,0.000) new_fields=(78.7,80.6) distance=5.56 loss=(-0.000,3.819) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=2.00 chaser_return=5.39 explorer_return=67.53 vision=(0.000,0.000) new_fields=(77.5,79.0) distance=4.95 loss=(-0.000,3.818) kl=0.0003
+update=16500 collisions=2.60 chaser_return=5.80 explorer_return=69.74 vision=(0.000,0.000) new_fields=(79.6,80.5) distance=5.16 loss=(-0.000,3.713) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=3.85 chaser_return=6.39 explorer_return=62.16 vision=(0.000,0.000) new_fields=(82.7,75.3) distance=5.26 loss=(-0.001,3.738) kl=0.0000
+update=17500 collisions=2.75 chaser_return=5.80 explorer_return=67.45 vision=(0.000,0.000) new_fields=(79.5,78.7) distance=4.74 loss=(-0.001,3.659) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=2.65 chaser_return=6.38 explorer_return=64.97 vision=(0.000,0.000) new_fields=(82.7,77.3) distance=5.41 loss=(-0.000,3.824) kl=0.0001
+update=18500 collisions=2.75 chaser_return=6.22 explorer_return=63.59 vision=(0.000,0.000) new_fields=(81.9,76.1) distance=5.82 loss=(-0.001,3.752) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=2.45 chaser_return=6.23 explorer_return=58.26 vision=(0.000,0.000) new_fields=(81.8,72.8) distance=6.27 loss=(-0.001,3.727) kl=0.0005
+update=19500 collisions=2.48 chaser_return=6.39 explorer_return=63.14 vision=(0.000,0.000) new_fields=(82.8,76.0) distance=5.31 loss=(-0.001,3.639) kl=0.0002
+update=20000 collisions=2.17 chaser_return=6.30 explorer_return=40.08 vision=(0.000,0.000) new_fields=(82.0,60.6) distance=4.87 loss=(-0.001,4.015) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7575.8237598598935, "metrics": {"approx_kl": 0.0005693047922270544, "chaser_grad_norm": 0.08362068422138691, "chaser_new_fields": 82.0250015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.3000006675720215, "chaser_subspace_norm": 109.71160888671875, "collisions_per_episode": 2.174999952316284, "entropy": 0.05865261983126402, "explorer_grad_norm": 0.6362885981798172, "explorer_new_fields": 60.57500076293945, "explorer_partner_vision": 0.0, "explorer_return": 40.07500076293945, "explorer_subspace_norm": 35.61513900756836, "final_distance": 4.870059967041016, "max_param_abs": 1.4202895164489746, "policy_loss": -0.0008471746835780891, "value_loss": 4.014994144439697}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01", "seed": 7, "task": "non_social", "unit_id": "non_social/seed_0007"}
+=== exited 2026-07-02T11:50:20.933771+00:00 returncode=0 ===

logs/non_social_seed_0008_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:58.683758+00:00 task=non_social seed=8 attempt=1 ===
+update=0001 collisions=1.20 chaser_return=-3.33 explorer_return=-0.46 vision=(0.000,0.000) new_fields=(33.5,33.2) distance=5.33 loss=(-0.001,3.079) kl=0.0000
+update=0500 collisions=2.85 chaser_return=4.65 explorer_return=57.33 vision=(0.000,0.000) new_fields=(73.8,72.1) distance=5.43 loss=(-0.002,4.429) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=4.15 chaser_return=4.26 explorer_return=52.98 vision=(0.000,0.000) new_fields=(72.0,69.2) distance=5.34 loss=(-0.002,4.264) kl=0.0003
+update=1500 collisions=3.73 chaser_return=4.90 explorer_return=48.40 vision=(0.000,0.000) new_fields=(75.3,66.3) distance=5.17 loss=(-0.001,3.912) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=3.42 chaser_return=4.84 explorer_return=50.35 vision=(0.000,0.000) new_fields=(75.0,67.4) distance=5.65 loss=(-0.001,4.110) kl=0.0001
+update=2500 collisions=3.12 chaser_return=5.36 explorer_return=50.54 vision=(0.000,0.000) new_fields=(77.4,67.5) distance=4.78 loss=(-0.002,3.743) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=2.98 chaser_return=5.52 explorer_return=49.71 vision=(0.000,0.000) new_fields=(78.2,67.1) distance=4.63 loss=(-0.001,3.709) kl=0.0000
+update=3500 collisions=2.90 chaser_return=5.25 explorer_return=49.53 vision=(0.000,0.000) new_fields=(77.0,66.9) distance=4.52 loss=(-0.001,3.540) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=2.88 chaser_return=5.54 explorer_return=50.58 vision=(0.000,0.000) new_fields=(78.4,67.6) distance=5.88 loss=(-0.001,3.740) kl=0.0003
+update=4500 collisions=2.45 chaser_return=5.63 explorer_return=51.59 vision=(0.000,0.000) new_fields=(78.9,68.2) distance=4.98 loss=(-0.001,3.567) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=2.50 chaser_return=5.56 explorer_return=52.30 vision=(0.000,0.000) new_fields=(78.3,68.9) distance=5.54 loss=(-0.000,3.617) kl=0.0005
+update=5500 collisions=2.17 chaser_return=6.34 explorer_return=50.35 vision=(0.000,0.000) new_fields=(82.3,67.5) distance=4.98 loss=(-0.000,3.635) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=1.60 chaser_return=6.55 explorer_return=48.03 vision=(0.000,0.000) new_fields=(83.1,65.9) distance=5.21 loss=(-0.000,3.480) kl=0.0006
+update=6500 collisions=1.33 chaser_return=6.65 explorer_return=52.64 vision=(0.000,0.000) new_fields=(83.6,69.0) distance=5.85 loss=(-0.001,3.727) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=2.28 chaser_return=6.36 explorer_return=53.69 vision=(0.000,0.000) new_fields=(82.3,69.6) distance=4.77 loss=(-0.001,3.508) kl=0.0003
+update=7500 collisions=1.85 chaser_return=6.62 explorer_return=52.15 vision=(0.000,0.000) new_fields=(83.5,68.8) distance=4.47 loss=(-0.000,3.314) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=2.03 chaser_return=6.33 explorer_return=53.91 vision=(0.000,0.000) new_fields=(82.2,69.7) distance=5.06 loss=(-0.000,3.409) kl=0.0002
+update=8500 collisions=2.20 chaser_return=6.65 explorer_return=52.60 vision=(0.000,0.000) new_fields=(83.9,68.9) distance=5.03 loss=(-0.001,3.339) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=1.93 chaser_return=6.39 explorer_return=52.38 vision=(0.000,0.000) new_fields=(82.5,68.8) distance=5.01 loss=(-0.001,3.413) kl=0.0002
+update=9500 collisions=2.28 chaser_return=6.09 explorer_return=50.35 vision=(0.000,0.000) new_fields=(81.0,67.5) distance=5.02 loss=(-0.001,3.551) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=2.33 chaser_return=6.30 explorer_return=50.24 vision=(0.000,0.000) new_fields=(82.3,67.2) distance=5.40 loss=(-0.001,3.602) kl=0.0001
+update=10500 collisions=1.43 chaser_return=6.74 explorer_return=52.67 vision=(0.000,0.000) new_fields=(84.1,68.8) distance=6.27 loss=(-0.001,3.465) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=2.03 chaser_return=6.50 explorer_return=53.54 vision=(0.000,0.000) new_fields=(83.2,69.3) distance=5.15 loss=(-0.002,3.284) kl=0.0008
+update=11500 collisions=1.88 chaser_return=6.59 explorer_return=49.86 vision=(0.000,0.000) new_fields=(83.5,67.2) distance=5.48 loss=(-0.000,3.423) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=2.23 chaser_return=6.77 explorer_return=53.65 vision=(0.000,0.000) new_fields=(84.5,69.8) distance=5.38 loss=(-0.000,3.443) kl=0.0002
+update=12500 collisions=1.40 chaser_return=6.42 explorer_return=54.66 vision=(0.000,0.000) new_fields=(82.3,70.2) distance=5.47 loss=(-0.001,3.402) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=2.35 chaser_return=6.18 explorer_return=52.98 vision=(0.000,0.000) new_fields=(81.5,69.3) distance=5.13 loss=(-0.001,3.602) kl=0.0003
+update=13500 collisions=1.90 chaser_return=6.92 explorer_return=55.64 vision=(0.000,0.000) new_fields=(85.2,70.9) distance=5.89 loss=(-0.001,3.192) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=2.65 chaser_return=6.15 explorer_return=54.44 vision=(0.000,0.000) new_fields=(81.5,70.2) distance=4.89 loss=(-0.001,3.364) kl=0.0000
+update=14500 collisions=1.68 chaser_return=6.32 explorer_return=54.36 vision=(0.000,0.000) new_fields=(82.0,70.0) distance=5.32 loss=(-0.000,3.215) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=2.85 chaser_return=5.25 explorer_return=56.65 vision=(0.000,0.000) new_fields=(76.8,71.7) distance=5.54 loss=(-0.000,3.466) kl=0.0003
+update=15500 collisions=2.62 chaser_return=5.09 explorer_return=55.15 vision=(0.000,0.000) new_fields=(76.0,70.8) distance=5.29 loss=(-0.001,3.261) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=2.88 chaser_return=5.08 explorer_return=54.48 vision=(0.000,0.000) new_fields=(76.1,70.2) distance=4.54 loss=(-0.001,3.437) kl=-0.0001
+update=16500 collisions=2.78 chaser_return=5.18 explorer_return=57.40 vision=(0.000,0.000) new_fields=(76.6,72.2) distance=5.48 loss=(-0.000,3.312) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=3.08 chaser_return=5.32 explorer_return=56.35 vision=(0.000,0.000) new_fields=(77.3,71.5) distance=4.99 loss=(-0.001,3.467) kl=0.0007
+update=17500 collisions=2.88 chaser_return=4.70 explorer_return=58.34 vision=(0.000,0.000) new_fields=(74.2,72.6) distance=5.35 loss=(-0.001,3.363) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=2.95 chaser_return=4.89 explorer_return=57.17 vision=(0.000,0.000) new_fields=(75.3,72.0) distance=4.61 loss=(-0.001,3.421) kl=0.0007
+update=18500 collisions=2.38 chaser_return=4.96 explorer_return=57.40 vision=(0.000,0.000) new_fields=(75.2,72.0) distance=4.20 loss=(-0.001,3.389) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=2.05 chaser_return=6.01 explorer_return=55.60 vision=(0.000,0.000) new_fields=(80.5,70.9) distance=5.28 loss=(-0.001,3.723) kl=0.0003
+update=19500 collisions=2.75 chaser_return=5.87 explorer_return=56.58 vision=(0.000,0.000) new_fields=(80.1,71.5) distance=4.57 loss=(-0.000,3.602) kl=0.0002
+update=20000 collisions=1.98 chaser_return=6.15 explorer_return=52.49 vision=(0.000,0.000) new_fields=(81.3,68.7) distance=4.52 loss=(-0.001,4.003) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7577.7527450309135, "metrics": {"approx_kl": 3.927184380664306e-06, "chaser_grad_norm": 0.07202051393687725, "chaser_new_fields": 81.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.145000457763672, "chaser_subspace_norm": 132.7053680419922, "collisions_per_episode": 1.975000023841858, "entropy": 0.13418130949139595, "explorer_grad_norm": 0.6797178164124489, "explorer_new_fields": 68.7249984741211, "explorer_partner_vision": 0.0, "explorer_return": 52.48749923706055, "explorer_subspace_norm": 61.86991500854492, "final_distance": 4.521096229553223, "max_param_abs": 1.4186153411865234, "policy_loss": -0.0009259324722812323, "value_loss": 4.002953112125397}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01", "seed": 8, "task": "non_social", "unit_id": "non_social/seed_0008"}
+=== exited 2026-07-02T11:50:20.938456+00:00 returncode=0 ===

logs/non_social_seed_0009_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:58.685349+00:00 task=non_social seed=9 attempt=1 ===
+update=0001 collisions=1.35 chaser_return=-3.14 explorer_return=3.10 vision=(0.000,0.000) new_fields=(34.4,35.7) distance=4.75 loss=(-0.001,3.000) kl=0.0001
+update=0500 collisions=3.05 chaser_return=4.38 explorer_return=45.29 vision=(0.000,0.000) new_fields=(72.6,64.0) distance=5.09 loss=(-0.002,4.753) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=3.45 chaser_return=5.18 explorer_return=48.40 vision=(0.000,0.000) new_fields=(76.7,66.2) distance=4.81 loss=(-0.001,4.740) kl=0.0001
+update=1500 collisions=2.58 chaser_return=4.79 explorer_return=52.19 vision=(0.000,0.000) new_fields=(74.6,68.7) distance=6.12 loss=(-0.002,4.563) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=3.05 chaser_return=5.24 explorer_return=53.50 vision=(0.000,0.000) new_fields=(76.8,69.6) distance=4.50 loss=(-0.001,4.315) kl=0.0003
+update=2500 collisions=2.92 chaser_return=5.47 explorer_return=49.00 vision=(0.000,0.000) new_fields=(78.1,66.5) distance=5.13 loss=(-0.001,4.246) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=2.30 chaser_return=6.44 explorer_return=54.74 vision=(0.000,0.000) new_fields=(82.8,70.3) distance=5.01 loss=(-0.001,3.880) kl=0.0001
+update=3500 collisions=2.58 chaser_return=5.34 explorer_return=56.95 vision=(0.000,0.000) new_fields=(77.2,71.7) distance=5.21 loss=(-0.001,4.040) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=1.88 chaser_return=5.17 explorer_return=56.46 vision=(0.000,0.000) new_fields=(76.3,71.5) distance=4.51 loss=(-0.001,3.939) kl=0.0004
+update=4500 collisions=1.93 chaser_return=6.03 explorer_return=58.00 vision=(0.000,0.000) new_fields=(80.8,72.5) distance=5.79 loss=(-0.001,3.977) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=2.95 chaser_return=5.56 explorer_return=52.71 vision=(0.000,0.000) new_fields=(78.6,69.2) distance=5.54 loss=(-0.001,4.011) kl=-0.0001
+update=5500 collisions=2.40 chaser_return=5.75 explorer_return=57.40 vision=(0.000,0.000) new_fields=(79.2,72.1) distance=4.86 loss=(-0.001,3.808) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=1.55 chaser_return=5.84 explorer_return=58.67 vision=(0.000,0.000) new_fields=(79.8,72.9) distance=4.76 loss=(-0.001,3.803) kl=0.0003
+update=6500 collisions=1.73 chaser_return=4.94 explorer_return=53.95 vision=(0.000,0.000) new_fields=(75.2,69.8) distance=5.27 loss=(-0.001,4.034) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=2.95 chaser_return=6.07 explorer_return=59.58 vision=(0.000,0.000) new_fields=(81.0,73.5) distance=4.63 loss=(-0.001,3.707) kl=-0.0000
+update=7500 collisions=2.62 chaser_return=6.30 explorer_return=60.29 vision=(0.000,0.000) new_fields=(82.2,74.0) distance=5.25 loss=(-0.001,3.637) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=2.17 chaser_return=6.12 explorer_return=56.54 vision=(0.000,0.000) new_fields=(81.1,71.6) distance=5.69 loss=(-0.001,3.499) kl=0.0001
+update=8500 collisions=2.17 chaser_return=4.77 explorer_return=60.85 vision=(0.000,0.000) new_fields=(74.6,74.3) distance=5.21 loss=(-0.001,3.824) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=3.08 chaser_return=5.23 explorer_return=57.44 vision=(0.000,0.000) new_fields=(76.9,72.2) distance=5.10 loss=(-0.001,3.890) kl=0.0005
+update=9500 collisions=1.83 chaser_return=6.21 explorer_return=53.76 vision=(0.000,0.000) new_fields=(81.5,69.6) distance=3.76 loss=(-0.001,3.725) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=2.50 chaser_return=6.37 explorer_return=46.90 vision=(0.000,0.000) new_fields=(82.3,65.2) distance=4.34 loss=(-0.001,3.911) kl=0.0005
+update=10500 collisions=2.88 chaser_return=6.33 explorer_return=50.58 vision=(0.000,0.000) new_fields=(82.4,67.6) distance=4.59 loss=(-0.001,4.059) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=3.03 chaser_return=6.39 explorer_return=55.53 vision=(0.000,0.000) new_fields=(82.8,71.0) distance=4.56 loss=(-0.001,3.804) kl=0.0002
+update=11500 collisions=2.60 chaser_return=6.76 explorer_return=57.21 vision=(0.000,0.000) new_fields=(84.4,72.0) distance=5.05 loss=(-0.000,3.809) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=2.67 chaser_return=6.65 explorer_return=61.86 vision=(0.000,0.000) new_fields=(84.1,75.3) distance=3.69 loss=(-0.001,3.711) kl=0.0001
+update=12500 collisions=2.67 chaser_return=6.31 explorer_return=57.33 vision=(0.000,0.000) new_fields=(82.3,72.2) distance=5.71 loss=(-0.001,3.847) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=2.80 chaser_return=6.88 explorer_return=53.95 vision=(0.000,0.000) new_fields=(85.2,69.9) distance=5.24 loss=(-0.001,4.037) kl=0.0001
+update=13500 collisions=2.50 chaser_return=6.67 explorer_return=59.24 vision=(0.000,0.000) new_fields=(84.2,73.2) distance=5.39 loss=(-0.001,3.924) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=2.45 chaser_return=6.14 explorer_return=58.38 vision=(0.000,0.000) new_fields=(81.3,72.8) distance=5.25 loss=(-0.001,4.016) kl=0.0005
+update=14500 collisions=2.23 chaser_return=5.58 explorer_return=57.10 vision=(0.000,0.000) new_fields=(78.4,71.8) distance=4.94 loss=(-0.002,4.241) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=2.60 chaser_return=6.81 explorer_return=57.55 vision=(0.000,0.000) new_fields=(84.6,72.6) distance=4.95 loss=(-0.001,3.872) kl=0.0004
+update=15500 collisions=2.65 chaser_return=7.34 explorer_return=59.09 vision=(0.000,0.000) new_fields=(87.6,73.4) distance=5.42 loss=(-0.001,3.516) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=2.28 chaser_return=7.18 explorer_return=58.53 vision=(0.000,0.000) new_fields=(86.7,73.0) distance=5.84 loss=(-0.001,3.693) kl=0.0005
+update=16500 collisions=2.40 chaser_return=6.90 explorer_return=57.14 vision=(0.000,0.000) new_fields=(85.0,72.0) distance=5.23 loss=(-0.001,3.883) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=2.03 chaser_return=6.84 explorer_return=60.89 vision=(0.000,0.000) new_fields=(84.8,74.5) distance=5.42 loss=(-0.000,3.668) kl=0.0001
+update=17500 collisions=3.33 chaser_return=6.56 explorer_return=57.21 vision=(0.000,0.000) new_fields=(83.6,72.1) distance=5.64 loss=(-0.001,3.709) kl=-0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=2.05 chaser_return=6.86 explorer_return=64.71 vision=(0.000,0.000) new_fields=(84.9,77.2) distance=5.80 loss=(-0.001,3.714) kl=0.0002
+update=18500 collisions=2.42 chaser_return=6.88 explorer_return=47.05 vision=(0.000,0.000) new_fields=(85.3,65.2) distance=5.64 loss=(-0.001,3.949) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=1.93 chaser_return=6.77 explorer_return=57.85 vision=(0.000,0.000) new_fields=(84.3,72.3) distance=5.45 loss=(-0.001,3.705) kl=0.0000
+update=19500 collisions=2.05 chaser_return=6.73 explorer_return=59.69 vision=(0.000,0.000) new_fields=(84.2,73.7) distance=5.33 loss=(-0.000,3.773) kl=-0.0001
+update=20000 collisions=2.73 chaser_return=6.34 explorer_return=47.46 vision=(0.000,0.000) new_fields=(82.6,65.7) distance=5.68 loss=(-0.001,3.992) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7570.80278072902, "metrics": {"approx_kl": 0.00014854547405664803, "chaser_grad_norm": 0.09859942272305489, "chaser_new_fields": 82.5999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.340001106262207, "chaser_subspace_norm": 98.81183624267578, "collisions_per_episode": 2.7250001430511475, "entropy": 0.07547229155898094, "explorer_grad_norm": 0.32836734503507614, "explorer_new_fields": 65.67500305175781, "explorer_partner_vision": 0.0, "explorer_return": 47.462501525878906, "explorer_subspace_norm": 26.826025009155273, "final_distance": 5.683727741241455, "max_param_abs": 1.2476426362991333, "policy_loss": -0.0011177757633866747, "value_loss": 3.991724371910095}, "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01", "seed": 9, "task": "non_social", "unit_id": "non_social/seed_0009"}
+=== exited 2026-07-02T11:50:20.942911+00:00 returncode=0 ===

logs/social_seed_0000_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.521136+00:00 task=social seed=0 attempt=1 ===
+update=0001 collisions=1.55 chaser_return=-1.18 explorer_return=-0.41 vision=(0.343,0.343) new_fields=(35.9,33.9) distance=5.38 loss=(-0.001,3.128) kl=-0.0000
+update=0500 collisions=13.38 chaser_return=15.03 explorer_return=35.67 vision=(0.875,0.875) new_fields=(54.5,64.4) distance=2.02 loss=(-0.006,5.704) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=13.23 chaser_return=14.47 explorer_return=33.01 vision=(0.868,0.868) new_fields=(52.7,63.3) distance=2.46 loss=(-0.008,5.706) kl=0.0012
+update=1500 collisions=14.07 chaser_return=15.30 explorer_return=35.14 vision=(0.844,0.844) new_fields=(52.6,65.2) distance=2.23 loss=(-0.006,5.844) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=15.45 chaser_return=16.17 explorer_return=29.43 vision=(0.933,0.933) new_fields=(49.6,62.4) distance=2.19 loss=(-0.005,5.605) kl=0.0012
+update=2500 collisions=16.85 chaser_return=17.93 explorer_return=26.55 vision=(0.924,0.924) new_fields=(50.5,60.9) distance=1.91 loss=(-0.005,5.639) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=15.62 chaser_return=16.34 explorer_return=31.66 vision=(0.934,0.934) new_fields=(49.5,63.9) distance=1.93 loss=(-0.005,5.336) kl=0.0006
+update=3500 collisions=20.50 chaser_return=21.50 explorer_return=25.25 vision=(0.940,0.940) new_fields=(49.3,62.2) distance=1.81 loss=(-0.005,5.557) kl=0.0015
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=21.05 chaser_return=22.48 explorer_return=22.31 vision=(0.954,0.954) new_fields=(50.6,60.0) distance=1.64 loss=(-0.005,5.912) kl=0.0010
+update=4500 collisions=16.83 chaser_return=17.33 explorer_return=28.96 vision=(0.930,0.930) new_fields=(48.0,62.9) distance=1.88 loss=(-0.006,5.590) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=20.12 chaser_return=21.21 explorer_return=28.25 vision=(0.931,0.931) new_fields=(50.1,63.9) distance=1.69 loss=(-0.004,5.518) kl=0.0004
+update=5500 collisions=16.23 chaser_return=16.47 explorer_return=20.86 vision=(0.930,0.930) new_fields=(47.4,57.3) distance=1.66 loss=(-0.004,5.827) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=20.88 chaser_return=21.59 explorer_return=23.86 vision=(0.955,0.955) new_fields=(48.8,62.4) distance=1.52 loss=(-0.003,5.077) kl=0.0006
+update=6500 collisions=18.62 chaser_return=19.72 explorer_return=27.16 vision=(0.949,0.949) new_fields=(50.3,63.4) distance=1.64 loss=(-0.003,5.263) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=18.43 chaser_return=19.00 explorer_return=26.48 vision=(0.932,0.932) new_fields=(47.8,61.7) distance=1.77 loss=(-0.006,5.619) kl=0.0003
+update=7500 collisions=21.52 chaser_return=22.36 explorer_return=25.56 vision=(0.940,0.940) new_fields=(48.4,63.3) distance=1.66 loss=(-0.004,5.634) kl=-0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=17.77 chaser_return=18.38 explorer_return=32.91 vision=(0.947,0.947) new_fields=(48.4,67.3) distance=1.71 loss=(-0.003,5.241) kl=0.0005
+update=8500 collisions=21.77 chaser_return=22.35 explorer_return=22.14 vision=(0.944,0.944) new_fields=(46.4,60.8) distance=1.55 loss=(-0.004,5.629) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=21.98 chaser_return=23.22 explorer_return=23.76 vision=(0.946,0.946) new_fields=(49.9,61.9) distance=1.76 loss=(-0.005,5.505) kl=0.0005
+update=9500 collisions=20.20 chaser_return=20.92 explorer_return=19.51 vision=(0.960,0.960) new_fields=(47.4,58.1) distance=1.76 loss=(-0.006,5.600) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=20.05 chaser_return=21.00 explorer_return=31.66 vision=(0.919,0.919) new_fields=(48.2,65.7) distance=2.10 loss=(-0.004,5.301) kl=0.0008
+update=10500 collisions=17.05 chaser_return=18.19 explorer_return=31.70 vision=(0.952,0.952) new_fields=(50.2,64.2) distance=1.99 loss=(-0.003,5.351) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=19.35 chaser_return=20.85 explorer_return=25.49 vision=(0.929,0.929) new_fields=(52.7,62.2) distance=1.61 loss=(-0.003,5.082) kl=0.0003
+update=11500 collisions=24.95 chaser_return=26.32 explorer_return=20.96 vision=(0.940,0.940) new_fields=(49.4,61.0) distance=1.45 loss=(-0.004,5.243) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=20.05 chaser_return=21.22 explorer_return=29.38 vision=(0.943,0.943) new_fields=(51.3,65.8) distance=1.68 loss=(-0.003,4.689) kl=0.0005
+update=12500 collisions=17.50 chaser_return=18.93 explorer_return=25.74 vision=(0.904,0.904) new_fields=(51.9,60.3) distance=1.66 loss=(-0.005,5.803) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=20.23 chaser_return=21.15 explorer_return=25.20 vision=(0.954,0.954) new_fields=(49.1,62.7) distance=1.55 loss=(-0.006,5.374) kl=0.0004
+update=13500 collisions=18.08 chaser_return=19.56 explorer_return=26.20 vision=(0.923,0.923) new_fields=(51.6,61.2) distance=1.71 loss=(-0.005,5.471) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=19.35 chaser_return=20.41 explorer_return=23.16 vision=(0.955,0.955) new_fields=(48.9,59.9) distance=1.65 loss=(-0.004,5.519) kl=0.0006
+update=14500 collisions=20.73 chaser_return=21.12 explorer_return=18.20 vision=(0.912,0.912) new_fields=(46.4,57.0) distance=1.97 loss=(-0.006,5.378) kl=0.0015
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=17.10 chaser_return=18.27 explorer_return=24.59 vision=(0.946,0.946) new_fields=(51.8,60.2) distance=1.70 loss=(-0.004,4.840) kl=0.0013
+update=15500 collisions=17.73 chaser_return=18.79 explorer_return=23.94 vision=(0.929,0.929) new_fields=(50.7,59.8) distance=1.70 loss=(-0.005,4.998) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=18.58 chaser_return=19.77 explorer_return=23.77 vision=(0.910,0.910) new_fields=(51.1,58.9) distance=1.97 loss=(-0.003,5.528) kl=0.0001
+update=16500 collisions=23.68 chaser_return=24.70 explorer_return=14.51 vision=(0.934,0.934) new_fields=(47.8,56.2) distance=1.58 loss=(-0.006,5.932) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=20.43 chaser_return=21.99 explorer_return=21.31 vision=(0.959,0.959) new_fields=(52.9,60.1) distance=1.48 loss=(-0.004,5.102) kl=0.0006
+update=17500 collisions=18.43 chaser_return=19.79 explorer_return=20.18 vision=(0.963,0.963) new_fields=(51.8,57.0) distance=1.54 loss=(-0.002,5.649) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=18.98 chaser_return=20.29 explorer_return=19.71 vision=(0.953,0.953) new_fields=(50.6,57.0) distance=1.57 loss=(-0.003,4.677) kl=-0.0003
+update=18500 collisions=19.60 chaser_return=20.53 explorer_return=22.33 vision=(0.970,0.970) new_fields=(50.8,60.7) distance=1.52 loss=(-0.003,5.175) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=19.85 chaser_return=21.41 explorer_return=21.23 vision=(0.938,0.938) new_fields=(52.0,58.1) distance=1.78 loss=(-0.002,4.832) kl=0.0003
+update=19500 collisions=17.95 chaser_return=18.80 explorer_return=24.95 vision=(0.943,0.943) new_fields=(50.0,61.0) distance=1.59 loss=(-0.004,5.202) kl=0.0002
+update=20000 collisions=21.45 chaser_return=23.26 explorer_return=31.04 vision=(0.949,0.949) new_fields=(52.9,65.6) distance=1.66 loss=(-0.004,4.788) kl=0.0028
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7633.112761525088, "metrics": {"approx_kl": 0.002805363851033249, "chaser_grad_norm": 0.188689436763525, "chaser_new_fields": 52.92499923706055, "chaser_partner_vision": 0.9490000605583191, "chaser_return": 23.260000228881836, "chaser_subspace_norm": 46.499481201171875, "collisions_per_episode": 21.450000762939453, "entropy": 0.1341038942337036, "explorer_grad_norm": 1.1695344895124435, "explorer_new_fields": 65.55000305175781, "explorer_partner_vision": 0.9490000605583191, "explorer_return": 31.037500381469727, "explorer_subspace_norm": 105.87859344482422, "final_distance": 1.655848741531372, "max_param_abs": 1.2022045850753784, "policy_loss": -0.003802475623324142, "value_loss": 4.788339614868164}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01", "seed": 0, "task": "social", "unit_id": "social/seed_0000"}
+=== exited 2026-07-02T09:43:38.591318+00:00 returncode=0 ===

logs/social_seed_0001_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.522668+00:00 task=social seed=1 attempt=1 ===
+update=0001 collisions=1.08 chaser_return=-1.94 explorer_return=3.50 vision=(0.320,0.320) new_fields=(34.5,36.2) distance=5.04 loss=(-0.001,2.967) kl=-0.0000
+update=0500 collisions=10.82 chaser_return=11.62 explorer_return=33.88 vision=(0.882,0.882) new_fields=(51.0,62.0) distance=2.11 loss=(-0.006,5.540) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=12.32 chaser_return=13.37 explorer_return=29.60 vision=(0.886,0.886) new_fields=(51.1,59.9) distance=1.95 loss=(-0.007,5.394) kl=0.0010
+update=1500 collisions=18.95 chaser_return=20.12 explorer_return=25.50 vision=(0.840,0.840) new_fields=(48.8,59.4) distance=2.35 loss=(-0.002,5.995) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=13.00 chaser_return=13.99 explorer_return=35.08 vision=(0.858,0.858) new_fields=(51.5,64.4) distance=2.15 loss=(-0.005,5.452) kl=0.0006
+update=2500 collisions=14.62 chaser_return=15.85 explorer_return=38.39 vision=(0.842,0.842) new_fields=(52.2,67.6) distance=2.44 loss=(-0.004,5.462) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=15.00 chaser_return=16.84 explorer_return=40.11 vision=(0.875,0.875) new_fields=(55.0,69.4) distance=1.66 loss=(-0.004,5.187) kl=0.0004
+update=3500 collisions=17.12 chaser_return=18.35 explorer_return=30.01 vision=(0.924,0.924) new_fields=(51.5,63.0) distance=1.99 loss=(-0.006,5.489) kl=0.0013
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=13.88 chaser_return=14.88 explorer_return=30.36 vision=(0.899,0.899) new_fields=(51.5,61.6) distance=1.93 loss=(-0.006,5.385) kl=0.0006
+update=4500 collisions=16.40 chaser_return=17.02 explorer_return=27.64 vision=(0.900,0.900) new_fields=(48.6,61.1) distance=1.71 loss=(-0.005,5.530) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=17.12 chaser_return=17.93 explorer_return=29.30 vision=(0.891,0.891) new_fields=(49.4,62.9) distance=1.77 loss=(-0.006,5.748) kl=0.0005
+update=5500 collisions=17.27 chaser_return=18.28 explorer_return=28.96 vision=(0.916,0.916) new_fields=(49.4,62.3) distance=2.05 loss=(-0.007,5.578) kl=0.0014
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=16.18 chaser_return=17.09 explorer_return=29.21 vision=(0.945,0.945) new_fields=(50.2,61.4) distance=1.69 loss=(-0.005,5.509) kl=0.0007
+update=6500 collisions=18.65 chaser_return=19.88 explorer_return=28.69 vision=(0.934,0.934) new_fields=(50.1,63.0) distance=1.69 loss=(-0.004,5.050) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=18.12 chaser_return=19.23 explorer_return=24.83 vision=(0.947,0.947) new_fields=(50.8,59.3) distance=1.47 loss=(-0.005,5.500) kl=0.0010
+update=7500 collisions=14.95 chaser_return=16.35 explorer_return=31.18 vision=(0.942,0.942) new_fields=(53.8,63.2) distance=1.81 loss=(-0.003,5.200) kl=-0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=21.08 chaser_return=22.54 explorer_return=27.33 vision=(0.953,0.953) new_fields=(52.5,64.6) distance=1.66 loss=(-0.002,5.103) kl=0.0002
+update=8500 collisions=23.15 chaser_return=24.65 explorer_return=20.10 vision=(0.942,0.942) new_fields=(50.9,59.7) distance=1.72 loss=(-0.005,5.367) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=19.08 chaser_return=18.64 explorer_return=18.46 vision=(0.939,0.939) new_fields=(43.0,57.1) distance=1.62 loss=(-0.005,5.450) kl=0.0004
+update=9500 collisions=24.18 chaser_return=24.82 explorer_return=10.48 vision=(0.945,0.945) new_fields=(45.9,53.7) distance=1.99 loss=(-0.003,5.661) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=19.08 chaser_return=19.90 explorer_return=20.08 vision=(0.940,0.940) new_fields=(49.4,58.6) distance=1.61 loss=(-0.004,5.291) kl=0.0007
+update=10500 collisions=23.60 chaser_return=25.38 explorer_return=24.52 vision=(0.954,0.954) new_fields=(52.6,64.8) distance=1.67 loss=(-0.004,4.451) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=20.75 chaser_return=22.15 explorer_return=21.68 vision=(0.941,0.941) new_fields=(51.9,59.5) distance=1.69 loss=(-0.005,5.132) kl=0.0006
+update=11500 collisions=24.48 chaser_return=26.40 explorer_return=19.29 vision=(0.946,0.946) new_fields=(53.0,61.4) distance=1.70 loss=(-0.004,4.278) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=25.12 chaser_return=26.09 explorer_return=14.09 vision=(0.941,0.941) new_fields=(45.2,55.6) distance=1.57 loss=(-0.005,5.136) kl=0.0003
+update=12500 collisions=25.00 chaser_return=26.43 explorer_return=18.61 vision=(0.958,0.958) new_fields=(49.6,60.0) distance=1.65 loss=(-0.005,5.023) kl=0.0018
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=16.98 chaser_return=17.30 explorer_return=18.76 vision=(0.916,0.916) new_fields=(46.1,54.7) distance=1.80 loss=(-0.004,5.845) kl=0.0006
+update=13500 collisions=23.68 chaser_return=23.58 explorer_return=14.25 vision=(0.899,0.899) new_fields=(41.4,54.0) distance=2.19 loss=(-0.004,6.881) kl=0.0013
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=20.45 chaser_return=21.04 explorer_return=21.56 vision=(0.942,0.942) new_fields=(46.6,59.7) distance=1.82 loss=(-0.004,6.012) kl=0.0006
+update=14500 collisions=20.18 chaser_return=21.29 explorer_return=25.68 vision=(0.950,0.950) new_fields=(49.2,61.6) distance=1.64 loss=(-0.005,5.376) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=20.58 chaser_return=22.01 explorer_return=20.19 vision=(0.961,0.961) new_fields=(50.5,57.2) distance=1.80 loss=(-0.005,5.727) kl=0.0016
+update=15500 collisions=18.43 chaser_return=19.35 explorer_return=21.41 vision=(0.909,0.909) new_fields=(50.2,58.9) distance=1.76 loss=(-0.005,5.373) kl=0.0013
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=18.23 chaser_return=19.38 explorer_return=19.41 vision=(0.952,0.952) new_fields=(50.3,57.0) distance=1.73 loss=(-0.004,5.053) kl=0.0007
+update=16500 collisions=17.15 chaser_return=18.02 explorer_return=22.27 vision=(0.935,0.935) new_fields=(49.4,57.2) distance=1.61 loss=(-0.004,5.534) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=19.30 chaser_return=20.44 explorer_return=25.29 vision=(0.952,0.952) new_fields=(48.6,61.2) distance=1.56 loss=(-0.004,4.820) kl=-0.0001
+update=17500 collisions=15.70 chaser_return=17.64 explorer_return=26.94 vision=(0.959,0.959) new_fields=(55.2,60.1) distance=1.57 loss=(-0.003,5.384) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=19.10 chaser_return=20.42 explorer_return=20.29 vision=(0.918,0.918) new_fields=(50.9,58.2) distance=1.51 loss=(-0.002,5.023) kl=0.0002
+update=18500 collisions=14.98 chaser_return=16.10 explorer_return=28.05 vision=(0.936,0.936) new_fields=(50.9,60.4) distance=2.05 loss=(-0.005,5.323) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=18.62 chaser_return=20.25 explorer_return=19.96 vision=(0.938,0.938) new_fields=(51.3,57.3) distance=1.63 loss=(-0.004,5.188) kl=0.0009
+update=19500 collisions=18.80 chaser_return=19.55 explorer_return=23.18 vision=(0.946,0.946) new_fields=(48.9,59.8) distance=1.64 loss=(-0.005,5.439) kl=0.0011
+update=20000 collisions=22.62 chaser_return=23.43 explorer_return=12.71 vision=(0.930,0.930) new_fields=(46.2,52.8) distance=1.79 loss=(-0.004,5.151) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7644.622009031009, "metrics": {"approx_kl": 0.0005562261293015069, "chaser_grad_norm": 0.6423081681132317, "chaser_new_fields": 46.20000076293945, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 23.4325008392334, "chaser_subspace_norm": 57.10129928588867, "collisions_per_episode": 22.625, "entropy": 0.14692536741495132, "explorer_grad_norm": 0.5817638859152794, "explorer_new_fields": 52.775001525878906, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 12.71250057220459, "explorer_subspace_norm": 122.46495819091797, "final_distance": 1.791115403175354, "max_param_abs": 1.3355721235275269, "policy_loss": -0.004476253691127141, "value_loss": 5.1513988971710205}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01", "seed": 1, "task": "social", "unit_id": "social/seed_0001"}
+=== exited 2026-07-02T09:43:48.618767+00:00 returncode=0 ===

logs/social_seed_0002_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.523612+00:00 task=social seed=2 attempt=1 ===
+update=0001 collisions=1.77 chaser_return=-1.36 explorer_return=4.43 vision=(0.351,0.351) new_fields=(33.7,37.0) distance=5.46 loss=(-0.001,3.191) kl=-0.0001
+update=0500 collisions=13.62 chaser_return=15.27 explorer_return=31.39 vision=(0.913,0.913) new_fields=(54.1,61.9) distance=2.02 loss=(-0.006,5.733) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=12.88 chaser_return=13.40 explorer_return=33.38 vision=(0.864,0.864) new_fields=(49.1,63.1) distance=2.16 loss=(-0.006,5.659) kl=0.0003
+update=1500 collisions=17.75 chaser_return=18.26 explorer_return=23.36 vision=(0.858,0.858) new_fields=(46.0,57.4) distance=2.22 loss=(-0.003,6.222) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=16.00 chaser_return=17.11 explorer_return=32.00 vision=(0.925,0.925) new_fields=(51.2,64.2) distance=1.80 loss=(-0.005,5.407) kl=0.0007
+update=2500 collisions=17.95 chaser_return=18.81 explorer_return=29.26 vision=(0.933,0.933) new_fields=(49.4,63.5) distance=1.71 loss=(-0.007,5.811) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=16.08 chaser_return=17.28 explorer_return=27.65 vision=(0.920,0.920) new_fields=(50.7,60.1) distance=1.92 loss=(-0.003,5.428) kl=0.0005
+update=3500 collisions=17.30 chaser_return=18.31 explorer_return=30.00 vision=(0.931,0.931) new_fields=(51.2,63.0) distance=1.75 loss=(-0.005,4.935) kl=0.0012
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=14.70 chaser_return=15.64 explorer_return=28.75 vision=(0.926,0.926) new_fields=(50.6,61.3) distance=1.90 loss=(-0.007,5.421) kl=0.0012
+update=4500 collisions=16.33 chaser_return=17.15 explorer_return=28.99 vision=(0.953,0.953) new_fields=(50.0,62.2) distance=1.75 loss=(-0.004,5.477) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=15.65 chaser_return=16.39 explorer_return=29.44 vision=(0.920,0.920) new_fields=(49.9,61.9) distance=1.92 loss=(-0.004,5.663) kl=0.0002
+update=5500 collisions=18.98 chaser_return=20.03 explorer_return=26.16 vision=(0.940,0.940) new_fields=(49.4,61.7) distance=1.63 loss=(-0.004,5.522) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=18.52 chaser_return=19.91 explorer_return=28.38 vision=(0.935,0.935) new_fields=(51.8,62.0) distance=1.78 loss=(-0.004,5.418) kl=0.0013
+update=6500 collisions=18.98 chaser_return=20.61 explorer_return=28.56 vision=(0.959,0.959) new_fields=(52.0,62.8) distance=1.77 loss=(-0.005,5.486) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=18.45 chaser_return=20.15 explorer_return=25.38 vision=(0.939,0.939) new_fields=(52.7,60.4) distance=1.61 loss=(-0.004,5.538) kl=0.0002
+update=7500 collisions=17.62 chaser_return=19.13 explorer_return=25.86 vision=(0.910,0.910) new_fields=(52.7,60.3) distance=1.91 loss=(-0.005,5.519) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=19.00 chaser_return=20.10 explorer_return=27.65 vision=(0.905,0.905) new_fields=(49.9,61.9) distance=1.72 loss=(-0.004,5.314) kl=0.0009
+update=8500 collisions=16.90 chaser_return=17.89 explorer_return=29.19 vision=(0.924,0.924) new_fields=(50.6,62.4) distance=1.69 loss=(-0.006,5.260) kl=0.0020
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=18.25 chaser_return=19.01 explorer_return=25.62 vision=(0.889,0.889) new_fields=(48.4,61.0) distance=1.78 loss=(-0.005,5.587) kl=0.0002
+update=9500 collisions=20.48 chaser_return=21.45 explorer_return=30.62 vision=(0.917,0.917) new_fields=(48.3,65.8) distance=1.77 loss=(-0.004,4.956) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=18.60 chaser_return=19.53 explorer_return=26.39 vision=(0.941,0.941) new_fields=(49.5,61.5) distance=1.78 loss=(-0.005,5.054) kl=0.0009
+update=10500 collisions=21.20 chaser_return=21.32 explorer_return=23.14 vision=(0.926,0.926) new_fields=(44.2,60.2) distance=1.72 loss=(-0.006,5.529) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=21.00 chaser_return=21.79 explorer_return=25.79 vision=(0.923,0.923) new_fields=(47.7,63.3) distance=1.73 loss=(-0.006,5.579) kl=-0.0002
+update=11500 collisions=20.48 chaser_return=22.33 explorer_return=28.60 vision=(0.945,0.945) new_fields=(53.2,65.0) distance=1.74 loss=(-0.005,4.895) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=19.62 chaser_return=20.70 explorer_return=20.62 vision=(0.915,0.915) new_fields=(49.9,58.0) distance=1.68 loss=(-0.003,5.800) kl=0.0005
+update=12500 collisions=19.58 chaser_return=20.77 explorer_return=21.89 vision=(0.958,0.958) new_fields=(49.8,58.9) distance=1.42 loss=(-0.004,5.128) kl=0.0015
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=20.93 chaser_return=22.59 explorer_return=23.69 vision=(0.930,0.930) new_fields=(53.0,61.8) distance=1.71 loss=(-0.004,5.010) kl=0.0006
+update=13500 collisions=20.80 chaser_return=21.57 explorer_return=21.24 vision=(0.934,0.934) new_fields=(47.9,58.9) distance=1.74 loss=(-0.005,5.529) kl=-0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=20.58 chaser_return=21.48 explorer_return=22.44 vision=(0.943,0.943) new_fields=(48.5,59.8) distance=1.82 loss=(-0.006,5.330) kl=0.0012
+update=14500 collisions=18.68 chaser_return=19.69 explorer_return=20.39 vision=(0.939,0.939) new_fields=(49.4,56.9) distance=1.82 loss=(-0.004,4.297) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=17.77 chaser_return=19.23 explorer_return=27.59 vision=(0.923,0.923) new_fields=(52.3,61.8) distance=1.95 loss=(-0.005,5.129) kl=0.0008
+update=15500 collisions=18.85 chaser_return=20.09 explorer_return=26.49 vision=(0.941,0.941) new_fields=(50.5,61.5) distance=1.80 loss=(-0.004,5.276) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=16.60 chaser_return=17.45 explorer_return=27.20 vision=(0.928,0.928) new_fields=(49.4,60.4) distance=1.57 loss=(-0.004,4.995) kl=0.0008
+update=16500 collisions=20.38 chaser_return=21.55 explorer_return=18.19 vision=(0.952,0.952) new_fields=(50.8,58.0) distance=1.53 loss=(-0.003,4.908) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=31.12 chaser_return=32.67 explorer_return=9.59 vision=(0.918,0.918) new_fields=(46.0,53.4) distance=1.52 loss=(-0.004,6.142) kl=0.0004
+update=17500 collisions=20.68 chaser_return=22.20 explorer_return=20.66 vision=(0.942,0.942) new_fields=(52.6,60.0) distance=1.58 loss=(-0.004,5.418) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=17.38 chaser_return=18.61 explorer_return=22.95 vision=(0.931,0.931) new_fields=(51.8,58.6) distance=1.69 loss=(-0.005,5.352) kl=0.0002
+update=18500 collisions=20.33 chaser_return=21.49 explorer_return=18.62 vision=(0.941,0.941) new_fields=(50.1,57.3) distance=1.71 loss=(-0.002,5.454) kl=0.0012
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=16.10 chaser_return=17.36 explorer_return=24.26 vision=(0.931,0.931) new_fields=(52.0,59.3) distance=1.59 loss=(-0.004,5.270) kl=0.0014
+update=19500 collisions=20.08 chaser_return=20.62 explorer_return=17.93 vision=(0.900,0.900) new_fields=(46.5,55.7) distance=1.88 loss=(-0.005,5.528) kl=0.0003
+update=20000 collisions=17.62 chaser_return=18.49 explorer_return=26.84 vision=(0.929,0.929) new_fields=(49.3,61.0) distance=1.66 loss=(-0.005,5.285) kl=0.0014
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7640.536813838058, "metrics": {"approx_kl": 0.0014266868641836794, "chaser_grad_norm": 0.20290611311793327, "chaser_new_fields": 49.29999923706055, "chaser_partner_vision": 0.9287500381469727, "chaser_return": 18.48750114440918, "chaser_subspace_norm": 48.7114143371582, "collisions_per_episode": 17.625, "entropy": 0.18204283714294434, "explorer_grad_norm": 0.6118505597114563, "explorer_new_fields": 60.95000076293945, "explorer_partner_vision": 0.9287500381469727, "explorer_return": 26.837499618530273, "explorer_subspace_norm": 104.41957092285156, "final_distance": 1.6572967767715454, "max_param_abs": 1.1167881488800049, "policy_loss": -0.005259661337355226, "value_loss": 5.285007953643799}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01", "seed": 2, "task": "social", "unit_id": "social/seed_0002"}
+=== exited 2026-07-02T09:43:48.626929+00:00 returncode=0 ===

logs/social_seed_0003_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.524464+00:00 task=social seed=3 attempt=1 ===
+update=0001 collisions=2.25 chaser_return=-1.33 explorer_return=-1.62 vision=(0.335,0.335) new_fields=(31.4,33.2) distance=4.66 loss=(-0.001,3.365) kl=0.0000
+update=0500 collisions=7.75 chaser_return=9.81 explorer_return=39.61 vision=(0.486,0.486) new_fields=(57.7,63.9) distance=4.32 loss=(-0.006,5.427) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=9.12 chaser_return=9.90 explorer_return=33.23 vision=(0.628,0.628) new_fields=(50.7,60.5) distance=3.20 loss=(-0.008,5.689) kl=0.0009
+update=1500 collisions=14.20 chaser_return=15.11 explorer_return=30.39 vision=(0.712,0.712) new_fields=(49.8,60.7) distance=3.09 loss=(-0.008,5.967) kl=0.0017
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=10.93 chaser_return=10.56 explorer_return=27.79 vision=(0.750,0.750) new_fields=(45.0,57.4) distance=2.43 loss=(-0.007,5.428) kl=0.0005
+update=2500 collisions=14.30 chaser_return=15.41 explorer_return=33.41 vision=(0.877,0.877) new_fields=(51.8,63.2) distance=1.97 loss=(-0.005,5.269) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=14.50 chaser_return=14.96 explorer_return=31.29 vision=(0.872,0.872) new_fields=(48.7,62.7) distance=2.60 loss=(-0.007,5.670) kl=0.0006
+update=3500 collisions=16.88 chaser_return=18.27 explorer_return=29.99 vision=(0.933,0.933) new_fields=(52.5,62.2) distance=2.05 loss=(-0.006,5.498) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=16.10 chaser_return=17.56 explorer_return=28.27 vision=(0.921,0.921) new_fields=(53.5,61.1) distance=2.00 loss=(-0.005,5.506) kl=0.0008
+update=4500 collisions=17.85 chaser_return=18.94 explorer_return=28.90 vision=(0.929,0.929) new_fields=(51.2,62.0) distance=1.85 loss=(-0.005,5.304) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=18.10 chaser_return=19.14 explorer_return=26.71 vision=(0.935,0.935) new_fields=(50.2,61.6) distance=2.15 loss=(-0.006,5.127) kl=0.0009
+update=5500 collisions=27.62 chaser_return=28.64 explorer_return=14.11 vision=(0.922,0.922) new_fields=(44.5,55.9) distance=1.75 loss=(-0.004,6.265) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=17.93 chaser_return=18.61 explorer_return=25.60 vision=(0.934,0.934) new_fields=(48.2,61.0) distance=1.95 loss=(-0.007,5.765) kl=0.0009
+update=6500 collisions=18.23 chaser_return=18.96 explorer_return=25.68 vision=(0.947,0.947) new_fields=(48.1,60.9) distance=1.87 loss=(-0.006,5.595) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=18.50 chaser_return=19.01 explorer_return=25.61 vision=(0.937,0.937) new_fields=(47.1,60.7) distance=1.81 loss=(-0.006,5.844) kl=0.0009
+update=7500 collisions=20.65 chaser_return=21.87 explorer_return=24.39 vision=(0.939,0.939) new_fields=(49.7,61.5) distance=1.63 loss=(-0.006,5.313) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=18.75 chaser_return=20.29 explorer_return=29.73 vision=(0.895,0.895) new_fields=(52.6,63.7) distance=1.94 loss=(-0.005,5.260) kl=0.0006
+update=8500 collisions=17.33 chaser_return=18.82 explorer_return=26.88 vision=(0.901,0.901) new_fields=(53.4,61.0) distance=1.99 loss=(-0.005,5.187) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=23.55 chaser_return=25.10 explorer_return=27.02 vision=(0.943,0.943) new_fields=(50.5,65.3) distance=1.62 loss=(-0.004,4.676) kl=0.0011
+update=9500 collisions=18.52 chaser_return=18.94 explorer_return=23.65 vision=(0.883,0.883) new_fields=(46.2,60.1) distance=1.90 loss=(-0.005,5.649) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=17.68 chaser_return=17.84 explorer_return=23.77 vision=(0.940,0.940) new_fields=(45.7,59.8) distance=2.10 loss=(-0.007,5.642) kl=0.0009
+update=10500 collisions=16.62 chaser_return=17.66 explorer_return=23.59 vision=(0.919,0.919) new_fields=(50.0,58.5) distance=1.86 loss=(-0.006,5.099) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=20.35 chaser_return=21.21 explorer_return=23.30 vision=(0.943,0.943) new_fields=(48.4,60.0) distance=1.65 loss=(-0.005,5.237) kl=0.0006
+update=11500 collisions=15.95 chaser_return=16.83 explorer_return=27.15 vision=(0.935,0.935) new_fields=(49.8,61.0) distance=1.67 loss=(-0.003,5.660) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=19.77 chaser_return=20.86 explorer_return=19.58 vision=(0.940,0.940) new_fields=(49.3,56.7) distance=1.87 loss=(-0.003,5.213) kl=0.0003
+update=12500 collisions=19.35 chaser_return=20.05 explorer_return=20.99 vision=(0.935,0.935) new_fields=(48.2,58.4) distance=1.66 loss=(-0.004,5.335) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=19.25 chaser_return=19.34 explorer_return=18.49 vision=(0.945,0.945) new_fields=(44.5,56.5) distance=2.06 loss=(-0.006,5.764) kl=0.0005
+update=13500 collisions=23.23 chaser_return=24.17 explorer_return=16.39 vision=(0.938,0.938) new_fields=(49.0,58.2) distance=1.79 loss=(-0.005,5.262) kl=0.0013
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=20.20 chaser_return=21.58 explorer_return=20.90 vision=(0.945,0.945) new_fields=(51.0,59.2) distance=1.94 loss=(-0.004,4.611) kl=0.0011
+update=14500 collisions=16.12 chaser_return=17.18 explorer_return=31.23 vision=(0.903,0.903) new_fields=(51.6,63.5) distance=2.14 loss=(-0.006,5.432) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=16.05 chaser_return=16.87 explorer_return=22.45 vision=(0.931,0.931) new_fields=(49.2,57.8) distance=1.76 loss=(-0.004,5.438) kl=0.0005
+update=15500 collisions=20.05 chaser_return=20.75 explorer_return=20.41 vision=(0.947,0.947) new_fields=(49.0,59.2) distance=1.56 loss=(-0.005,5.328) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=12.62 chaser_return=12.98 explorer_return=29.41 vision=(0.926,0.926) new_fields=(48.0,59.9) distance=1.76 loss=(-0.004,6.173) kl=0.0011
+update=16500 collisions=20.08 chaser_return=19.93 explorer_return=14.32 vision=(0.928,0.928) new_fields=(43.2,53.1) distance=1.80 loss=(-0.005,5.542) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=20.80 chaser_return=21.68 explorer_return=18.69 vision=(0.940,0.940) new_fields=(48.4,57.7) distance=1.56 loss=(-0.004,5.817) kl=0.0008
+update=17500 collisions=15.78 chaser_return=16.62 explorer_return=28.48 vision=(0.920,0.920) new_fields=(49.9,61.6) distance=1.89 loss=(-0.005,5.507) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=19.35 chaser_return=20.32 explorer_return=21.25 vision=(0.959,0.959) new_fields=(49.2,59.3) distance=1.61 loss=(-0.006,5.306) kl=0.0008
+update=18500 collisions=23.23 chaser_return=24.16 explorer_return=19.05 vision=(0.934,0.934) new_fields=(48.2,58.4) distance=1.72 loss=(-0.005,5.771) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=21.02 chaser_return=22.85 explorer_return=26.49 vision=(0.946,0.946) new_fields=(52.6,63.4) distance=1.78 loss=(-0.004,4.452) kl=-0.0002
+update=19500 collisions=18.98 chaser_return=19.46 explorer_return=19.11 vision=(0.929,0.929) new_fields=(46.8,56.8) distance=1.70 loss=(-0.004,5.291) kl=0.0002
+update=20000 collisions=20.20 chaser_return=21.88 explorer_return=22.89 vision=(0.930,0.930) new_fields=(52.5,59.3) distance=1.49 loss=(-0.003,5.231) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7642.245397494058, "metrics": {"approx_kl": 0.000801136315581541, "chaser_grad_norm": 0.38031933456659317, "chaser_new_fields": 52.5, "chaser_partner_vision": 0.9297500252723694, "chaser_return": 21.880001068115234, "chaser_subspace_norm": 52.41925811767578, "collisions_per_episode": 20.200000762939453, "entropy": 0.12773804366588593, "explorer_grad_norm": 0.23243309557437897, "explorer_new_fields": 59.29999923706055, "explorer_partner_vision": 0.9297500252723694, "explorer_return": 22.887500762939453, "explorer_subspace_norm": 119.99281311035156, "final_distance": 1.4895528554916382, "max_param_abs": 1.4659936428070068, "policy_loss": -0.0027591757117768623, "value_loss": 5.230631113052368}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01", "seed": 3, "task": "social", "unit_id": "social/seed_0003"}
+=== exited 2026-07-02T09:43:48.634923+00:00 returncode=0 ===

logs/social_seed_0004_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T07:36:16.525323+00:00 task=social seed=4 attempt=1 ===
+update=0001 collisions=1.70 chaser_return=-0.83 explorer_return=0.68 vision=(0.378,0.378) new_fields=(36.7,34.6) distance=5.05 loss=(-0.001,3.205) kl=-0.0000
+update=0500 collisions=6.38 chaser_return=8.84 explorer_return=42.36 vision=(0.460,0.460) new_fields=(60.2,65.0) distance=4.21 loss=(-0.006,5.577) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=12.53 chaser_return=14.09 explorer_return=34.83 vision=(0.582,0.582) new_fields=(53.2,62.9) distance=3.61 loss=(-0.007,6.011) kl=0.0005
+update=1500 collisions=12.03 chaser_return=12.55 explorer_return=33.24 vision=(0.725,0.725) new_fields=(49.0,62.2) distance=3.14 loss=(-0.007,5.755) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=14.45 chaser_return=15.05 explorer_return=28.31 vision=(0.835,0.835) new_fields=(48.7,59.5) distance=2.22 loss=(-0.006,5.787) kl=0.0008
+update=2500 collisions=13.50 chaser_return=14.67 explorer_return=30.70 vision=(0.835,0.835) new_fields=(52.0,60.9) distance=2.56 loss=(-0.006,5.782) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=21.48 chaser_return=21.84 explorer_return=23.56 vision=(0.914,0.914) new_fields=(44.4,59.7) distance=1.73 loss=(-0.003,6.184) kl=0.0003
+update=3500 collisions=16.05 chaser_return=16.70 explorer_return=27.25 vision=(0.880,0.880) new_fields=(49.1,60.4) distance=1.93 loss=(-0.006,5.322) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=16.80 chaser_return=17.41 explorer_return=24.06 vision=(0.840,0.840) new_fields=(48.5,58.0) distance=1.94 loss=(-0.005,5.411) kl=0.0008
+update=4500 collisions=17.80 chaser_return=18.23 explorer_return=25.06 vision=(0.918,0.918) new_fields=(46.4,60.2) distance=1.92 loss=(-0.006,5.698) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=21.48 chaser_return=22.33 explorer_return=19.74 vision=(0.930,0.930) new_fields=(47.9,58.2) distance=1.64 loss=(-0.005,5.707) kl=0.0011
+update=5500 collisions=18.27 chaser_return=18.97 explorer_return=24.34 vision=(0.936,0.936) new_fields=(48.4,59.5) distance=1.70 loss=(-0.008,5.519) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=17.05 chaser_return=18.25 explorer_return=29.60 vision=(0.894,0.894) new_fields=(50.0,61.9) distance=1.89 loss=(-0.006,4.938) kl=0.0008
+update=6500 collisions=16.48 chaser_return=17.04 explorer_return=26.06 vision=(0.933,0.933) new_fields=(48.3,60.5) distance=1.66 loss=(-0.006,5.170) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=16.55 chaser_return=17.50 explorer_return=27.26 vision=(0.923,0.923) new_fields=(49.6,61.3) distance=1.84 loss=(-0.006,5.630) kl=0.0012
+update=7500 collisions=22.00 chaser_return=23.08 explorer_return=20.34 vision=(0.933,0.933) new_fields=(48.1,59.0) distance=1.63 loss=(-0.005,5.723) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=17.62 chaser_return=18.70 explorer_return=28.41 vision=(0.923,0.923) new_fields=(50.4,63.1) distance=1.45 loss=(-0.004,5.002) kl=0.0004
+update=8500 collisions=20.15 chaser_return=21.70 explorer_return=24.11 vision=(0.941,0.941) new_fields=(51.0,61.0) distance=1.79 loss=(-0.005,4.883) kl=0.0016
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=19.48 chaser_return=20.04 explorer_return=23.96 vision=(0.930,0.930) new_fields=(48.3,61.5) distance=2.08 loss=(-0.005,5.764) kl=0.0013
+update=9500 collisions=18.62 chaser_return=20.44 explorer_return=30.91 vision=(0.945,0.945) new_fields=(53.0,64.1) distance=1.67 loss=(-0.004,4.688) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=14.10 chaser_return=14.21 explorer_return=34.23 vision=(0.926,0.926) new_fields=(46.1,63.9) distance=2.01 loss=(-0.003,5.754) kl=0.0003
+update=10500 collisions=16.65 chaser_return=17.77 explorer_return=24.40 vision=(0.947,0.947) new_fields=(51.1,59.2) distance=1.61 loss=(-0.005,5.539) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=20.60 chaser_return=22.02 explorer_return=26.36 vision=(0.933,0.933) new_fields=(50.0,61.6) distance=1.73 loss=(-0.003,5.189) kl=0.0009
+update=11500 collisions=16.68 chaser_return=17.77 explorer_return=35.30 vision=(0.954,0.954) new_fields=(50.3,67.7) distance=1.71 loss=(-0.005,4.762) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=16.88 chaser_return=17.46 explorer_return=28.49 vision=(0.935,0.935) new_fields=(47.8,62.2) distance=1.98 loss=(-0.005,5.271) kl=0.0008
+update=12500 collisions=18.55 chaser_return=19.29 explorer_return=24.05 vision=(0.943,0.943) new_fields=(48.9,60.7) distance=1.67 loss=(-0.003,5.154) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=16.02 chaser_return=17.47 explorer_return=30.34 vision=(0.947,0.947) new_fields=(52.1,64.1) distance=1.71 loss=(-0.004,5.049) kl=0.0000
+update=13500 collisions=16.60 chaser_return=18.08 explorer_return=33.99 vision=(0.947,0.947) new_fields=(52.9,66.9) distance=1.72 loss=(-0.003,4.745) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=18.45 chaser_return=19.59 explorer_return=27.59 vision=(0.952,0.952) new_fields=(50.1,60.9) distance=2.01 loss=(-0.005,5.542) kl=0.0007
+update=14500 collisions=20.52 chaser_return=21.76 explorer_return=19.69 vision=(0.938,0.938) new_fields=(50.0,57.7) distance=1.65 loss=(-0.003,5.719) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=18.10 chaser_return=19.14 explorer_return=23.23 vision=(0.951,0.951) new_fields=(48.9,58.3) distance=1.80 loss=(-0.004,5.447) kl=0.0004
+update=15500 collisions=9.18 chaser_return=10.16 explorer_return=22.06 vision=(0.944,0.944) new_fields=(52.6,53.2) distance=2.47 loss=(-0.002,6.393) kl=-0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=25.98 chaser_return=26.84 explorer_return=8.82 vision=(0.940,0.940) new_fields=(44.2,51.6) distance=1.72 loss=(-0.005,6.720) kl=0.0008
+update=16500 collisions=20.18 chaser_return=20.67 explorer_return=17.27 vision=(0.945,0.945) new_fields=(45.8,55.1) distance=1.45 loss=(-0.005,5.433) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=21.25 chaser_return=22.05 explorer_return=17.04 vision=(0.941,0.941) new_fields=(48.5,55.2) distance=1.96 loss=(-0.005,5.166) kl=0.0008
+update=17500 collisions=17.95 chaser_return=18.55 explorer_return=21.01 vision=(0.935,0.935) new_fields=(47.4,56.9) distance=1.55 loss=(-0.005,5.306) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=23.90 chaser_return=24.92 explorer_return=11.62 vision=(0.945,0.945) new_fields=(47.5,53.8) distance=1.61 loss=(-0.002,4.897) kl=0.0001
+update=18500 collisions=18.73 chaser_return=19.05 explorer_return=15.71 vision=(0.951,0.951) new_fields=(45.0,54.1) distance=1.60 loss=(-0.005,5.166) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=15.98 chaser_return=16.43 explorer_return=18.85 vision=(0.886,0.886) new_fields=(47.9,54.1) distance=2.02 loss=(-0.005,5.096) kl=0.0011
+update=19500 collisions=20.58 chaser_return=21.39 explorer_return=27.01 vision=(0.934,0.934) new_fields=(47.5,62.9) distance=1.57 loss=(-0.003,4.533) kl=0.0000
+update=20000 collisions=19.12 chaser_return=19.42 explorer_return=19.30 vision=(0.923,0.923) new_fields=(45.6,56.9) distance=1.83 loss=(-0.005,5.130) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7640.287686598953, "metrics": {"approx_kl": 0.000236154562574864, "chaser_grad_norm": 0.38722849637269974, "chaser_new_fields": 45.57500076293945, "chaser_partner_vision": 0.9225000143051147, "chaser_return": 19.41750144958496, "chaser_subspace_norm": 58.249393463134766, "collisions_per_episode": 19.125, "entropy": 0.19220001995563507, "explorer_grad_norm": 0.3825383335351944, "explorer_new_fields": 56.85000228881836, "explorer_partner_vision": 0.9225000143051147, "explorer_return": 19.30000114440918, "explorer_subspace_norm": 90.54417419433594, "final_distance": 1.830224633216858, "max_param_abs": 1.0662928819656372, "policy_loss": -0.004796283046713867, "value_loss": 5.130034923553467}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01", "seed": 4, "task": "social", "unit_id": "social/seed_0004"}
+=== exited 2026-07-02T09:43:48.659872+00:00 returncode=0 ===

logs/social_seed_0005_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:18.366281+00:00 task=social seed=5 attempt=1 ===
+update=0001 collisions=1.30 chaser_return=-1.92 explorer_return=1.51 vision=(0.305,0.305) new_fields=(33.5,35.0) distance=5.48 loss=(-0.001,3.134) kl=-0.0000
+update=0500 collisions=13.03 chaser_return=14.07 explorer_return=31.12 vision=(0.923,0.923) new_fields=(51.4,61.0) distance=1.82 loss=(-0.003,5.539) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=15.38 chaser_return=16.13 explorer_return=29.09 vision=(0.922,0.922) new_fields=(49.2,61.4) distance=1.61 loss=(-0.006,5.701) kl=0.0003
+update=1500 collisions=13.53 chaser_return=13.99 explorer_return=32.34 vision=(0.912,0.912) new_fields=(49.2,62.6) distance=1.86 loss=(-0.005,5.412) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=15.95 chaser_return=17.07 explorer_return=32.62 vision=(0.934,0.934) new_fields=(51.7,64.7) distance=1.64 loss=(-0.005,5.396) kl=0.0003
+update=2500 collisions=15.82 chaser_return=16.84 explorer_return=30.44 vision=(0.862,0.862) new_fields=(50.9,62.5) distance=2.38 loss=(-0.005,5.692) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=15.38 chaser_return=16.06 explorer_return=32.61 vision=(0.898,0.898) new_fields=(50.3,65.2) distance=2.14 loss=(-0.005,5.627) kl=0.0009
+update=3500 collisions=13.75 chaser_return=14.53 explorer_return=34.17 vision=(0.894,0.894) new_fields=(50.6,64.2) distance=2.00 loss=(-0.006,5.376) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=16.38 chaser_return=17.21 explorer_return=29.11 vision=(0.927,0.927) new_fields=(49.6,62.7) distance=1.93 loss=(-0.006,5.346) kl=0.0010
+update=4500 collisions=16.95 chaser_return=17.44 explorer_return=30.81 vision=(0.936,0.936) new_fields=(47.6,63.5) distance=1.81 loss=(-0.006,5.516) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=18.20 chaser_return=19.41 explorer_return=29.93 vision=(0.951,0.951) new_fields=(50.6,63.7) distance=1.94 loss=(-0.003,5.560) kl=0.0009
+update=5500 collisions=20.88 chaser_return=21.36 explorer_return=25.74 vision=(0.924,0.924) new_fields=(46.5,62.2) distance=1.70 loss=(-0.004,5.574) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=17.48 chaser_return=18.21 explorer_return=29.09 vision=(0.934,0.934) new_fields=(48.8,62.5) distance=1.75 loss=(-0.004,5.393) kl=0.0003
+update=6500 collisions=18.25 chaser_return=19.27 explorer_return=27.54 vision=(0.933,0.933) new_fields=(50.6,62.3) distance=1.53 loss=(-0.004,5.698) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=23.20 chaser_return=24.22 explorer_return=15.24 vision=(0.959,0.959) new_fields=(47.4,56.0) distance=1.47 loss=(-0.002,5.525) kl=0.0004
+update=7500 collisions=17.20 chaser_return=18.11 explorer_return=26.11 vision=(0.945,0.945) new_fields=(49.0,60.7) distance=1.74 loss=(-0.006,5.427) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=18.75 chaser_return=19.70 explorer_return=28.11 vision=(0.936,0.936) new_fields=(50.2,63.5) distance=1.72 loss=(-0.004,5.190) kl=0.0005
+update=8500 collisions=21.68 chaser_return=23.25 explorer_return=25.19 vision=(0.955,0.955) new_fields=(53.2,64.1) distance=1.50 loss=(-0.002,5.458) kl=-0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=23.80 chaser_return=25.32 explorer_return=21.84 vision=(0.933,0.933) new_fields=(51.4,62.1) distance=1.66 loss=(-0.003,5.290) kl=0.0003
+update=9500 collisions=22.35 chaser_return=23.66 explorer_return=21.25 vision=(0.949,0.949) new_fields=(50.8,60.6) distance=1.57 loss=(-0.004,5.181) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=20.27 chaser_return=21.33 explorer_return=20.11 vision=(0.938,0.938) new_fields=(48.9,58.4) distance=1.55 loss=(-0.005,5.533) kl=0.0007
+update=10500 collisions=21.23 chaser_return=22.04 explorer_return=23.27 vision=(0.920,0.920) new_fields=(47.7,60.8) distance=1.86 loss=(-0.005,5.430) kl=0.0016
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=22.68 chaser_return=23.52 explorer_return=18.65 vision=(0.935,0.935) new_fields=(46.9,57.2) distance=1.66 loss=(-0.006,5.623) kl=0.0013
+update=11500 collisions=22.75 chaser_return=23.92 explorer_return=21.09 vision=(0.944,0.944) new_fields=(48.9,61.4) distance=1.77 loss=(-0.004,5.128) kl=0.0014
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=20.45 chaser_return=20.45 explorer_return=21.41 vision=(0.925,0.925) new_fields=(43.4,59.6) distance=1.70 loss=(-0.005,5.113) kl=0.0011
+update=12500 collisions=21.90 chaser_return=23.28 explorer_return=24.59 vision=(0.938,0.938) new_fields=(50.8,63.5) distance=1.39 loss=(-0.004,5.261) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=20.43 chaser_return=21.03 explorer_return=19.77 vision=(0.931,0.931) new_fields=(46.5,58.6) distance=1.57 loss=(-0.004,5.448) kl=0.0010
+update=13500 collisions=19.02 chaser_return=20.70 explorer_return=30.41 vision=(0.954,0.954) new_fields=(53.2,65.9) distance=1.36 loss=(-0.004,5.111) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=22.40 chaser_return=22.98 explorer_return=17.06 vision=(0.948,0.948) new_fields=(46.5,58.0) distance=1.60 loss=(-0.006,5.585) kl=0.0009
+update=14500 collisions=19.43 chaser_return=20.25 explorer_return=25.08 vision=(0.940,0.940) new_fields=(48.6,61.8) distance=1.63 loss=(-0.003,5.608) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=18.88 chaser_return=19.23 explorer_return=22.76 vision=(0.926,0.926) new_fields=(46.3,59.2) distance=1.96 loss=(-0.004,5.758) kl=0.0005
+update=15500 collisions=22.75 chaser_return=23.76 explorer_return=22.18 vision=(0.934,0.934) new_fields=(47.8,59.2) distance=1.78 loss=(-0.005,5.995) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=20.80 chaser_return=22.15 explorer_return=23.83 vision=(0.963,0.963) new_fields=(50.9,61.7) distance=1.65 loss=(-0.003,5.587) kl=0.0006
+update=16500 collisions=21.12 chaser_return=22.20 explorer_return=24.19 vision=(0.952,0.952) new_fields=(48.8,61.3) distance=1.58 loss=(-0.004,5.014) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=19.77 chaser_return=21.96 explorer_return=31.31 vision=(0.951,0.951) new_fields=(55.0,66.8) distance=1.62 loss=(-0.003,4.686) kl=0.0004
+update=17500 collisions=18.55 chaser_return=19.74 explorer_return=24.80 vision=(0.950,0.950) new_fields=(49.8,61.0) distance=1.75 loss=(-0.005,4.887) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=19.40 chaser_return=20.70 explorer_return=22.91 vision=(0.959,0.959) new_fields=(50.5,59.9) distance=1.65 loss=(-0.003,5.135) kl=0.0003
+update=18500 collisions=22.45 chaser_return=23.47 explorer_return=24.76 vision=(0.948,0.948) new_fields=(48.6,62.8) distance=1.47 loss=(-0.002,5.429) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=20.02 chaser_return=21.27 explorer_return=23.01 vision=(0.936,0.936) new_fields=(50.7,58.5) distance=1.65 loss=(-0.004,5.677) kl=0.0002
+update=19500 collisions=16.45 chaser_return=17.51 explorer_return=29.38 vision=(0.922,0.922) new_fields=(50.2,62.4) distance=1.90 loss=(-0.005,4.853) kl=0.0006
+update=20000 collisions=18.58 chaser_return=20.12 explorer_return=20.77 vision=(0.953,0.953) new_fields=(51.5,58.6) distance=1.58 loss=(-0.003,4.460) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7556.884778791922, "metrics": {"approx_kl": 0.0005827252787078407, "chaser_grad_norm": 0.3728112056851387, "chaser_new_fields": 51.47500228881836, "chaser_partner_vision": 0.95250004529953, "chaser_return": 20.117502212524414, "chaser_subspace_norm": 50.84516143798828, "collisions_per_episode": 18.575000762939453, "entropy": 0.13566195219755173, "explorer_grad_norm": 0.6062372475862503, "explorer_new_fields": 58.57500076293945, "explorer_partner_vision": 0.95250004529953, "explorer_return": 20.774999618530273, "explorer_subspace_norm": 104.22787475585938, "final_distance": 1.5808099508285522, "max_param_abs": 1.3590010404586792, "policy_loss": -0.003127762281341262, "value_loss": 4.4600828886032104}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01", "seed": 5, "task": "social", "unit_id": "social/seed_0005"}
+=== exited 2026-07-02T11:49:20.760552+00:00 returncode=0 ===

logs/social_seed_0006_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:38.590438+00:00 task=social seed=6 attempt=1 ===
+update=0001 collisions=2.45 chaser_return=-0.59 explorer_return=-1.69 vision=(0.423,0.423) new_fields=(34.0,33.5) distance=4.61 loss=(-0.001,3.144) kl=-0.0001
+update=0500 collisions=10.60 chaser_return=11.59 explorer_return=32.08 vision=(0.866,0.866) new_fields=(51.8,60.5) distance=2.10 loss=(-0.006,5.398) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=10.78 chaser_return=11.69 explorer_return=34.73 vision=(0.892,0.892) new_fields=(51.7,62.7) distance=2.16 loss=(-0.007,5.382) kl=0.0011
+update=1500 collisions=16.58 chaser_return=18.04 explorer_return=25.56 vision=(0.913,0.913) new_fields=(51.8,59.0) distance=2.27 loss=(-0.005,5.727) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=14.12 chaser_return=15.67 explorer_return=35.53 vision=(0.914,0.914) new_fields=(53.6,64.7) distance=1.87 loss=(-0.007,5.695) kl=0.0009
+update=2500 collisions=15.82 chaser_return=16.51 explorer_return=28.52 vision=(0.895,0.895) new_fields=(48.5,61.3) distance=1.95 loss=(-0.006,5.511) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=15.85 chaser_return=16.76 explorer_return=30.05 vision=(0.934,0.934) new_fields=(50.0,62.4) distance=1.69 loss=(-0.005,5.600) kl=0.0003
+update=3500 collisions=20.73 chaser_return=21.29 explorer_return=23.00 vision=(0.905,0.905) new_fields=(45.4,58.6) distance=1.98 loss=(-0.004,5.874) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=17.58 chaser_return=17.99 explorer_return=24.01 vision=(0.906,0.906) new_fields=(46.7,59.8) distance=2.17 loss=(-0.006,5.487) kl=0.0007
+update=4500 collisions=17.98 chaser_return=18.74 explorer_return=26.29 vision=(0.922,0.922) new_fields=(49.6,61.5) distance=1.86 loss=(-0.006,5.626) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=16.38 chaser_return=17.31 explorer_return=32.90 vision=(0.915,0.915) new_fields=(50.2,65.5) distance=1.74 loss=(-0.006,5.136) kl=0.0010
+update=5500 collisions=17.20 chaser_return=18.22 explorer_return=29.71 vision=(0.930,0.930) new_fields=(50.0,63.4) distance=1.70 loss=(-0.005,5.252) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=19.43 chaser_return=20.94 explorer_return=24.40 vision=(0.941,0.941) new_fields=(51.2,61.4) distance=1.44 loss=(-0.006,5.647) kl=0.0006
+update=6500 collisions=19.00 chaser_return=20.03 explorer_return=26.34 vision=(0.946,0.946) new_fields=(49.3,62.2) distance=1.66 loss=(-0.008,5.644) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=18.08 chaser_return=19.50 explorer_return=27.93 vision=(0.935,0.935) new_fields=(51.8,63.1) distance=1.62 loss=(-0.005,5.577) kl=0.0023
+update=7500 collisions=16.90 chaser_return=17.54 explorer_return=27.73 vision=(0.872,0.872) new_fields=(48.5,61.9) distance=1.98 loss=(-0.006,5.703) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=17.93 chaser_return=18.60 explorer_return=23.91 vision=(0.919,0.919) new_fields=(47.8,59.9) distance=1.55 loss=(-0.004,5.587) kl=0.0002
+update=8500 collisions=15.35 chaser_return=16.16 explorer_return=28.50 vision=(0.913,0.913) new_fields=(49.8,61.7) distance=2.01 loss=(-0.006,5.374) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=22.93 chaser_return=23.25 explorer_return=19.39 vision=(0.958,0.958) new_fields=(43.9,59.0) distance=1.75 loss=(-0.005,5.772) kl=0.0009
+update=9500 collisions=18.52 chaser_return=19.30 explorer_return=23.54 vision=(0.916,0.916) new_fields=(49.0,59.7) distance=2.08 loss=(-0.006,5.346) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=17.85 chaser_return=19.39 explorer_return=26.76 vision=(0.943,0.943) new_fields=(53.2,61.2) distance=1.79 loss=(-0.005,5.027) kl=0.0010
+update=10500 collisions=17.80 chaser_return=18.59 explorer_return=20.38 vision=(0.945,0.945) new_fields=(48.1,57.2) distance=1.84 loss=(-0.006,5.263) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=19.60 chaser_return=21.08 explorer_return=27.39 vision=(0.935,0.935) new_fields=(51.7,63.4) distance=1.69 loss=(-0.003,4.567) kl=0.0003
+update=11500 collisions=19.65 chaser_return=20.52 explorer_return=26.12 vision=(0.933,0.933) new_fields=(48.6,62.0) distance=1.70 loss=(-0.005,5.235) kl=0.0016
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=26.08 chaser_return=27.74 explorer_return=18.04 vision=(0.958,0.958) new_fields=(50.5,59.3) distance=1.64 loss=(-0.005,5.639) kl=0.0009
+update=12500 collisions=20.52 chaser_return=22.68 explorer_return=34.31 vision=(0.956,0.956) new_fields=(55.2,68.9) distance=1.74 loss=(-0.004,5.144) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=27.95 chaser_return=29.96 explorer_return=18.00 vision=(0.955,0.955) new_fields=(52.7,61.4) distance=1.54 loss=(-0.005,5.030) kl=0.0010
+update=13500 collisions=18.88 chaser_return=19.42 explorer_return=19.46 vision=(0.938,0.938) new_fields=(46.8,57.2) distance=1.64 loss=(-0.005,5.378) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=19.75 chaser_return=21.11 explorer_return=25.74 vision=(0.955,0.955) new_fields=(51.2,61.6) distance=1.57 loss=(-0.003,5.500) kl=-0.0000
+update=14500 collisions=19.60 chaser_return=21.12 explorer_return=23.38 vision=(0.951,0.951) new_fields=(51.9,60.0) distance=1.67 loss=(-0.004,5.271) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=17.85 chaser_return=18.62 explorer_return=25.52 vision=(0.957,0.957) new_fields=(47.9,61.5) distance=2.02 loss=(-0.003,4.485) kl=0.0002
+update=15500 collisions=18.95 chaser_return=20.85 explorer_return=26.70 vision=(0.951,0.951) new_fields=(52.9,62.6) distance=1.57 loss=(-0.003,4.751) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=24.48 chaser_return=26.91 explorer_return=25.29 vision=(0.939,0.939) new_fields=(53.7,63.5) distance=1.58 loss=(-0.004,4.981) kl=0.0008
+update=16500 collisions=32.12 chaser_return=33.47 explorer_return=7.62 vision=(0.945,0.945) new_fields=(44.3,53.1) distance=1.35 loss=(-0.003,6.302) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=18.62 chaser_return=19.92 explorer_return=23.94 vision=(0.933,0.933) new_fields=(51.0,60.4) distance=2.08 loss=(-0.005,5.160) kl=0.0005
+update=17500 collisions=24.20 chaser_return=25.67 explorer_return=18.79 vision=(0.941,0.941) new_fields=(48.7,59.0) distance=1.74 loss=(-0.004,5.314) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=22.60 chaser_return=23.50 explorer_return=17.56 vision=(0.955,0.955) new_fields=(47.3,57.0) distance=1.60 loss=(-0.005,5.722) kl=0.0001
+update=18500 collisions=18.80 chaser_return=19.51 explorer_return=21.90 vision=(0.952,0.952) new_fields=(47.6,57.5) distance=1.68 loss=(-0.004,5.387) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=17.95 chaser_return=18.95 explorer_return=24.99 vision=(0.941,0.941) new_fields=(50.0,59.9) distance=1.50 loss=(-0.005,4.928) kl=0.0005
+update=19500 collisions=18.70 chaser_return=19.60 explorer_return=21.35 vision=(0.947,0.947) new_fields=(49.1,58.4) distance=1.58 loss=(-0.004,5.470) kl=0.0009
+update=20000 collisions=19.80 chaser_return=21.02 explorer_return=22.49 vision=(0.925,0.925) new_fields=(49.9,59.4) distance=1.64 loss=(-0.004,4.686) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7558.831456203014, "metrics": {"approx_kl": 0.00025015939701367174, "chaser_grad_norm": 0.31481391191482544, "chaser_new_fields": 49.900001525878906, "chaser_partner_vision": 0.9252500534057617, "chaser_return": 21.020002365112305, "chaser_subspace_norm": 49.21955108642578, "collisions_per_episode": 19.80000114440918, "entropy": 0.1560382954776287, "explorer_grad_norm": 0.6951030343770981, "explorer_new_fields": 59.42499923706055, "explorer_partner_vision": 0.9252500534057617, "explorer_return": 22.48750114440918, "explorer_subspace_norm": 113.72798919677734, "final_distance": 1.6350990533828735, "max_param_abs": 1.6354095935821533, "policy_loss": -0.0035889741121788443, "value_loss": 4.686129331588745}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01", "seed": 6, "task": "social", "unit_id": "social/seed_0006"}
+=== exited 2026-07-02T11:49:40.902414+00:00 returncode=0 ===

logs/social_seed_0007_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:48.618172+00:00 task=social seed=7 attempt=1 ===
+update=0001 collisions=1.38 chaser_return=-1.75 explorer_return=0.99 vision=(0.339,0.339) new_fields=(34.1,34.7) distance=5.21 loss=(-0.001,3.165) kl=0.0000
+update=0500 collisions=8.05 chaser_return=10.61 explorer_return=38.71 vision=(0.504,0.504) new_fields=(60.2,63.5) distance=3.50 loss=(-0.006,5.447) kl=0.0012
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=11.03 chaser_return=13.36 explorer_return=34.98 vision=(0.539,0.539) new_fields=(58.0,62.2) distance=3.88 loss=(-0.006,6.057) kl=0.0001
+update=1500 collisions=12.70 chaser_return=14.61 explorer_return=33.54 vision=(0.561,0.561) new_fields=(55.0,62.1) distance=3.62 loss=(-0.004,5.964) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=13.03 chaser_return=13.92 explorer_return=31.73 vision=(0.689,0.689) new_fields=(50.5,61.6) distance=2.93 loss=(-0.007,5.827) kl=0.0001
+update=2500 collisions=13.05 chaser_return=13.67 explorer_return=29.69 vision=(0.764,0.764) new_fields=(49.1,60.1) distance=2.39 loss=(-0.007,5.747) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=16.40 chaser_return=17.32 explorer_return=29.81 vision=(0.874,0.874) new_fields=(50.9,62.9) distance=1.84 loss=(-0.005,5.488) kl=-0.0001
+update=3500 collisions=16.30 chaser_return=17.11 explorer_return=29.11 vision=(0.945,0.945) new_fields=(49.8,61.5) distance=1.91 loss=(-0.005,5.188) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=17.27 chaser_return=17.98 explorer_return=25.36 vision=(0.911,0.911) new_fields=(49.0,60.2) distance=2.09 loss=(-0.005,5.741) kl=-0.0002
+update=4500 collisions=17.38 chaser_return=18.45 explorer_return=29.66 vision=(0.940,0.940) new_fields=(49.7,63.3) distance=1.76 loss=(-0.006,5.378) kl=0.0013
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=17.55 chaser_return=17.68 explorer_return=22.56 vision=(0.928,0.928) new_fields=(45.8,58.2) distance=1.77 loss=(-0.006,5.665) kl=0.0003
+update=5500 collisions=15.15 chaser_return=16.34 explorer_return=32.12 vision=(0.945,0.945) new_fields=(52.0,64.8) distance=1.70 loss=(-0.004,5.231) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=17.43 chaser_return=18.39 explorer_return=25.70 vision=(0.920,0.920) new_fields=(49.5,59.9) distance=1.98 loss=(-0.006,5.880) kl=0.0006
+update=6500 collisions=18.08 chaser_return=18.49 explorer_return=20.99 vision=(0.941,0.941) new_fields=(47.1,57.9) distance=1.81 loss=(-0.007,5.635) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=19.62 chaser_return=21.00 explorer_return=24.49 vision=(0.954,0.954) new_fields=(50.6,60.3) distance=1.67 loss=(-0.003,5.471) kl=0.0006
+update=7500 collisions=18.20 chaser_return=19.45 explorer_return=34.46 vision=(0.946,0.946) new_fields=(51.3,66.5) distance=1.81 loss=(-0.004,5.055) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=19.08 chaser_return=19.99 explorer_return=26.34 vision=(0.922,0.922) new_fields=(48.8,62.2) distance=1.77 loss=(-0.004,5.324) kl=0.0007
+update=8500 collisions=18.35 chaser_return=18.99 explorer_return=25.39 vision=(0.953,0.953) new_fields=(48.6,61.3) distance=1.76 loss=(-0.006,5.502) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=15.73 chaser_return=16.87 explorer_return=28.80 vision=(0.930,0.930) new_fields=(51.5,61.9) distance=1.64 loss=(-0.006,5.099) kl=0.0006
+update=9500 collisions=22.27 chaser_return=22.89 explorer_return=18.89 vision=(0.914,0.914) new_fields=(46.2,57.4) distance=1.82 loss=(-0.004,5.842) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=17.08 chaser_return=18.38 explorer_return=32.36 vision=(0.955,0.955) new_fields=(52.2,64.9) distance=1.55 loss=(-0.002,5.039) kl=-0.0003
+update=10500 collisions=20.27 chaser_return=20.30 explorer_return=20.38 vision=(0.951,0.951) new_fields=(43.8,59.0) distance=1.77 loss=(-0.005,5.443) kl=0.0019
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=24.60 chaser_return=25.43 explorer_return=19.15 vision=(0.947,0.947) new_fields=(46.5,59.5) distance=1.52 loss=(-0.006,5.679) kl=0.0012
+update=11500 collisions=20.80 chaser_return=21.65 explorer_return=19.48 vision=(0.934,0.934) new_fields=(48.4,57.9) distance=1.78 loss=(-0.004,5.281) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=23.35 chaser_return=24.62 explorer_return=20.00 vision=(0.910,0.910) new_fields=(50.2,60.8) distance=1.63 loss=(-0.003,5.122) kl=0.0003
+update=12500 collisions=20.23 chaser_return=21.25 explorer_return=23.21 vision=(0.943,0.943) new_fields=(50.1,61.1) distance=1.90 loss=(-0.003,4.965) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=19.23 chaser_return=19.65 explorer_return=10.25 vision=(0.953,0.953) new_fields=(46.8,50.4) distance=1.93 loss=(-0.003,5.366) kl=-0.0003
+update=13500 collisions=20.02 chaser_return=20.71 explorer_return=23.76 vision=(0.942,0.942) new_fields=(47.7,60.1) distance=1.68 loss=(-0.005,5.469) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=19.80 chaser_return=20.08 explorer_return=24.21 vision=(0.942,0.942) new_fields=(44.8,60.4) distance=2.14 loss=(-0.005,5.046) kl=0.0013
+update=14500 collisions=16.70 chaser_return=16.76 explorer_return=28.01 vision=(0.911,0.911) new_fields=(46.1,62.4) distance=1.93 loss=(-0.004,5.335) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=20.40 chaser_return=21.94 explorer_return=24.59 vision=(0.941,0.941) new_fields=(51.5,61.7) distance=1.65 loss=(-0.004,5.087) kl=0.0004
+update=15500 collisions=23.65 chaser_return=24.59 explorer_return=17.49 vision=(0.930,0.930) new_fields=(47.6,58.2) distance=1.64 loss=(-0.004,4.988) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=20.38 chaser_return=21.50 explorer_return=22.20 vision=(0.936,0.936) new_fields=(49.5,60.7) distance=1.61 loss=(-0.005,4.797) kl=0.0009
+update=16500 collisions=18.48 chaser_return=19.23 explorer_return=24.99 vision=(0.919,0.919) new_fields=(48.9,61.4) distance=1.68 loss=(-0.006,5.468) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=19.12 chaser_return=20.75 explorer_return=27.89 vision=(0.926,0.926) new_fields=(54.1,61.5) distance=1.80 loss=(-0.002,5.051) kl=0.0004
+update=17500 collisions=19.75 chaser_return=20.71 explorer_return=20.98 vision=(0.935,0.935) new_fields=(48.5,58.0) distance=1.58 loss=(-0.006,5.685) kl=0.0016
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=18.50 chaser_return=19.34 explorer_return=21.08 vision=(0.943,0.943) new_fields=(48.0,58.2) distance=1.59 loss=(-0.003,4.998) kl=-0.0002
+update=18500 collisions=21.93 chaser_return=22.19 explorer_return=9.43 vision=(0.932,0.932) new_fields=(42.9,50.0) distance=1.68 loss=(-0.004,5.543) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=18.02 chaser_return=18.05 explorer_return=18.01 vision=(0.945,0.945) new_fields=(45.5,55.5) distance=1.52 loss=(-0.003,5.345) kl=0.0006
+update=19500 collisions=18.75 chaser_return=19.64 explorer_return=21.66 vision=(0.925,0.925) new_fields=(49.4,58.0) distance=1.95 loss=(-0.005,5.504) kl=0.0005
+update=20000 collisions=23.20 chaser_return=24.29 explorer_return=17.15 vision=(0.947,0.947) new_fields=(46.9,55.9) distance=1.69 loss=(-0.003,4.716) kl=0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7587.021667060093, "metrics": {"approx_kl": 0.00013653355120357702, "chaser_grad_norm": 0.9264960587024689, "chaser_new_fields": 46.92499923706055, "chaser_partner_vision": 0.9470000267028809, "chaser_return": 24.285001754760742, "chaser_subspace_norm": 67.47857666015625, "collisions_per_episode": 23.200000762939453, "entropy": 0.1407805271446705, "explorer_grad_norm": 0.5267750471830368, "explorer_new_fields": 55.900001525878906, "explorer_partner_vision": 0.9470000267028809, "explorer_return": 17.149999618530273, "explorer_subspace_norm": 89.71617889404297, "final_distance": 1.689988374710083, "max_param_abs": 1.2198518514633179, "policy_loss": -0.003363337086513596, "value_loss": 4.716352343559265}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01", "seed": 7, "task": "social", "unit_id": "social/seed_0007"}
+=== exited 2026-07-02T11:50:20.928628+00:00 returncode=0 ===

logs/social_seed_0008_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:58.682914+00:00 task=social seed=8 attempt=1 ===
+update=0001 collisions=1.20 chaser_return=-2.02 explorer_return=-1.06 vision=(0.313,0.313) new_fields=(33.5,33.2) distance=5.32 loss=(-0.001,3.142) kl=0.0000
+update=0500 collisions=6.83 chaser_return=9.10 explorer_return=39.33 vision=(0.550,0.550) new_fields=(59.3,63.1) distance=4.29 loss=(-0.006,5.485) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=11.62 chaser_return=12.99 explorer_return=35.20 vision=(0.729,0.729) new_fields=(53.3,62.7) distance=2.64 loss=(-0.004,5.647) kl=0.0006
+update=1500 collisions=12.00 chaser_return=12.21 explorer_return=29.58 vision=(0.801,0.801) new_fields=(47.3,59.5) distance=1.95 loss=(-0.007,5.451) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=10.90 chaser_return=11.75 explorer_return=35.08 vision=(0.868,0.868) new_fields=(51.8,63.0) distance=2.37 loss=(-0.006,5.021) kl=0.0001
+update=2500 collisions=12.75 chaser_return=13.52 explorer_return=31.94 vision=(0.903,0.903) new_fields=(50.3,62.0) distance=1.81 loss=(-0.007,5.398) kl=0.0012
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=13.90 chaser_return=14.72 explorer_return=30.65 vision=(0.892,0.892) new_fields=(49.7,61.9) distance=2.06 loss=(-0.006,5.417) kl=0.0010
+update=3500 collisions=14.68 chaser_return=15.10 explorer_return=31.76 vision=(0.869,0.869) new_fields=(47.8,63.0) distance=2.22 loss=(-0.006,5.582) kl=0.0018
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=15.60 chaser_return=16.34 explorer_return=29.99 vision=(0.898,0.898) new_fields=(49.5,62.4) distance=2.12 loss=(-0.005,5.858) kl=0.0008
+update=4500 collisions=15.15 chaser_return=15.73 explorer_return=28.71 vision=(0.889,0.889) new_fields=(47.9,61.0) distance=1.94 loss=(-0.004,5.343) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=17.00 chaser_return=17.99 explorer_return=28.76 vision=(0.889,0.889) new_fields=(49.3,61.7) distance=1.86 loss=(-0.005,5.814) kl=0.0008
+update=5500 collisions=19.68 chaser_return=21.23 explorer_return=32.04 vision=(0.915,0.915) new_fields=(50.9,65.5) distance=1.87 loss=(-0.005,5.346) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=18.52 chaser_return=19.54 explorer_return=22.49 vision=(0.933,0.933) new_fields=(48.6,58.5) distance=1.71 loss=(-0.006,5.770) kl=0.0014
+update=6500 collisions=19.00 chaser_return=19.54 explorer_return=23.56 vision=(0.918,0.918) new_fields=(48.0,60.2) distance=1.92 loss=(-0.005,5.712) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=18.23 chaser_return=19.06 explorer_return=23.27 vision=(0.894,0.894) new_fields=(48.4,59.2) distance=1.81 loss=(-0.005,5.607) kl=0.0015
+update=7500 collisions=16.70 chaser_return=17.57 explorer_return=30.11 vision=(0.930,0.930) new_fields=(49.8,62.9) distance=1.84 loss=(-0.006,5.406) kl=0.0011
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=15.40 chaser_return=16.61 explorer_return=31.62 vision=(0.938,0.938) new_fields=(51.7,63.1) distance=1.77 loss=(-0.006,5.285) kl=0.0009
+update=8500 collisions=15.23 chaser_return=15.96 explorer_return=24.44 vision=(0.909,0.909) new_fields=(49.7,58.6) distance=1.98 loss=(-0.005,5.613) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=20.58 chaser_return=21.17 explorer_return=20.41 vision=(0.947,0.947) new_fields=(48.1,59.1) distance=1.81 loss=(-0.006,5.469) kl=0.0005
+update=9500 collisions=22.60 chaser_return=23.70 explorer_return=18.12 vision=(0.939,0.939) new_fields=(49.2,58.2) distance=1.60 loss=(-0.005,5.487) kl=0.0021
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=20.18 chaser_return=20.66 explorer_return=20.43 vision=(0.932,0.932) new_fields=(46.4,58.2) distance=1.64 loss=(-0.004,5.325) kl=0.0005
+update=10500 collisions=20.40 chaser_return=21.83 explorer_return=25.45 vision=(0.945,0.945) new_fields=(51.4,61.8) distance=1.43 loss=(-0.003,5.363) kl=-0.0001
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=18.85 chaser_return=19.87 explorer_return=24.01 vision=(0.936,0.936) new_fields=(49.7,61.0) distance=1.72 loss=(-0.005,5.403) kl=0.0013
+update=11500 collisions=16.52 chaser_return=17.88 explorer_return=22.02 vision=(0.944,0.944) new_fields=(51.5,58.3) distance=1.72 loss=(-0.005,5.244) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=19.05 chaser_return=20.25 explorer_return=22.56 vision=(0.951,0.951) new_fields=(50.1,58.5) distance=1.99 loss=(-0.005,5.343) kl=0.0011
+update=12500 collisions=18.18 chaser_return=18.68 explorer_return=26.49 vision=(0.892,0.892) new_fields=(47.2,62.0) distance=1.89 loss=(-0.006,5.339) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=17.65 chaser_return=19.17 explorer_return=25.33 vision=(0.938,0.938) new_fields=(52.0,60.5) distance=1.71 loss=(-0.004,5.004) kl=0.0008
+update=13500 collisions=18.50 chaser_return=18.90 explorer_return=17.77 vision=(0.944,0.944) new_fields=(47.2,55.9) distance=1.94 loss=(-0.003,5.405) kl=0.0008
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=17.95 chaser_return=19.27 explorer_return=19.81 vision=(0.926,0.926) new_fields=(51.1,57.3) distance=1.83 loss=(-0.003,5.690) kl=0.0006
+update=14500 collisions=17.95 chaser_return=18.61 explorer_return=23.60 vision=(0.912,0.912) new_fields=(47.5,59.5) distance=1.85 loss=(-0.002,4.620) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=27.30 chaser_return=28.75 explorer_return=18.36 vision=(0.904,0.904) new_fields=(47.0,57.9) distance=1.60 loss=(0.002,5.836) kl=0.0139
+update=15500 collisions=17.55 chaser_return=18.63 explorer_return=27.77 vision=(0.902,0.902) new_fields=(50.2,62.0) distance=1.71 loss=(-0.003,4.693) kl=-0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=18.27 chaser_return=19.75 explorer_return=24.79 vision=(0.914,0.914) new_fields=(51.7,59.4) distance=1.78 loss=(-0.002,4.726) kl=0.0004
+update=16500 collisions=18.73 chaser_return=20.18 explorer_return=19.39 vision=(0.906,0.906) new_fields=(51.2,55.9) distance=1.95 loss=(-0.004,5.000) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=15.98 chaser_return=17.19 explorer_return=28.15 vision=(0.916,0.916) new_fields=(51.8,61.7) distance=1.78 loss=(-0.004,5.103) kl=0.0010
+update=17500 collisions=20.90 chaser_return=22.00 explorer_return=18.19 vision=(0.922,0.922) new_fields=(48.6,57.2) distance=1.83 loss=(-0.004,5.165) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=19.90 chaser_return=21.10 explorer_return=21.99 vision=(0.878,0.878) new_fields=(49.7,59.7) distance=1.80 loss=(-0.005,5.162) kl=0.0011
+update=18500 collisions=22.43 chaser_return=23.93 explorer_return=18.33 vision=(0.921,0.921) new_fields=(50.5,57.9) distance=1.66 loss=(-0.004,5.221) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=22.15 chaser_return=22.85 explorer_return=15.20 vision=(0.855,0.855) new_fields=(45.8,54.5) distance=2.01 loss=(-0.004,5.898) kl=0.0006
+update=19500 collisions=17.43 chaser_return=18.17 explorer_return=26.64 vision=(0.907,0.907) new_fields=(48.9,61.5) distance=2.11 loss=(-0.004,4.408) kl=0.0004
+update=20000 collisions=19.68 chaser_return=20.85 explorer_return=23.00 vision=(0.921,0.921) new_fields=(50.1,60.5) distance=1.83 loss=(-0.005,4.698) kl=0.0023
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7584.432566124015, "metrics": {"approx_kl": 0.0022941239069333985, "chaser_grad_norm": 0.43133843690156937, "chaser_new_fields": 50.125, "chaser_partner_vision": 0.9205000400543213, "chaser_return": 20.852500915527344, "chaser_subspace_norm": 52.702186584472656, "collisions_per_episode": 19.67500114440918, "entropy": 0.15483620017766953, "explorer_grad_norm": 0.3490636646747589, "explorer_new_fields": 60.525001525878906, "explorer_partner_vision": 0.9205000400543213, "explorer_return": 23.0, "explorer_subspace_norm": 101.96681213378906, "final_distance": 1.8305482864379883, "max_param_abs": 1.1802449226379395, "policy_loss": -0.00493374926503698, "value_loss": 4.697651386260986}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01", "seed": 8, "task": "social", "unit_id": "social/seed_0008"}
+=== exited 2026-07-02T11:50:30.952355+00:00 returncode=0 ===

logs/social_seed_0009_attempt_01.log

@@ -0,0 +1,65 @@
+
+=== started 2026-07-02T09:43:58.684577+00:00 task=social seed=9 attempt=1 ===
+update=0001 collisions=1.38 chaser_return=-1.62 explorer_return=2.15 vision=(0.315,0.315) new_fields=(34.5,35.5) distance=4.80 loss=(-0.001,3.074) kl=0.0001
+update=0500 collisions=6.12 chaser_return=8.67 explorer_return=40.76 vision=(0.493,0.493) new_fields=(60.8,63.8) distance=4.03 loss=(-0.006,5.433) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_001000.safetensors
+update=1000 collisions=11.48 chaser_return=12.86 explorer_return=34.00 vision=(0.626,0.626) new_fields=(53.2,62.2) distance=3.38 loss=(-0.007,5.897) kl=0.0007
+update=1500 collisions=11.40 chaser_return=12.35 explorer_return=35.54 vision=(0.765,0.765) new_fields=(51.5,63.3) distance=2.84 loss=(-0.008,5.701) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_002000.safetensors
+update=2000 collisions=10.35 chaser_return=11.32 explorer_return=33.70 vision=(0.746,0.746) new_fields=(51.9,61.4) distance=2.18 loss=(-0.005,5.547) kl=0.0002
+update=2500 collisions=12.48 chaser_return=13.42 explorer_return=31.74 vision=(0.672,0.672) new_fields=(50.6,60.9) distance=2.45 loss=(-0.004,5.377) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_003000.safetensors
+update=3000 collisions=13.95 chaser_return=13.94 explorer_return=31.23 vision=(0.810,0.810) new_fields=(46.0,61.7) distance=2.36 loss=(-0.008,5.507) kl=0.0004
+update=3500 collisions=19.62 chaser_return=20.07 explorer_return=24.30 vision=(0.900,0.900) new_fields=(46.2,59.9) distance=1.85 loss=(-0.005,5.836) kl=0.0004
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_004000.safetensors
+update=4000 collisions=18.08 chaser_return=19.12 explorer_return=28.11 vision=(0.922,0.922) new_fields=(50.0,62.3) distance=1.74 loss=(-0.005,5.758) kl=0.0006
+update=4500 collisions=19.73 chaser_return=20.26 explorer_return=24.51 vision=(0.879,0.879) new_fields=(46.0,59.5) distance=2.18 loss=(-0.004,6.040) kl=0.0012
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_005000.safetensors
+update=5000 collisions=18.12 chaser_return=18.81 explorer_return=25.65 vision=(0.912,0.912) new_fields=(48.0,60.9) distance=1.92 loss=(-0.005,5.578) kl=0.0008
+update=5500 collisions=16.12 chaser_return=17.06 explorer_return=32.31 vision=(0.903,0.903) new_fields=(50.2,65.0) distance=1.90 loss=(-0.005,5.384) kl=0.0006
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_006000.safetensors
+update=6000 collisions=16.58 chaser_return=17.38 explorer_return=31.04 vision=(0.928,0.928) new_fields=(49.6,63.4) distance=1.78 loss=(-0.006,5.411) kl=0.0009
+update=6500 collisions=15.32 chaser_return=16.75 explorer_return=31.74 vision=(0.926,0.926) new_fields=(53.5,63.4) distance=1.94 loss=(-0.007,5.397) kl=0.0015
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_007000.safetensors
+update=7000 collisions=15.05 chaser_return=16.26 explorer_return=26.51 vision=(0.930,0.930) new_fields=(52.4,59.9) distance=1.82 loss=(-0.006,5.528) kl=0.0013
+update=7500 collisions=19.75 chaser_return=21.01 explorer_return=26.34 vision=(0.923,0.923) new_fields=(50.7,61.7) distance=1.75 loss=(-0.003,5.391) kl=0.0002
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_008000.safetensors
+update=8000 collisions=18.25 chaser_return=18.58 explorer_return=21.50 vision=(0.936,0.936) new_fields=(46.8,58.3) distance=1.65 loss=(-0.006,5.550) kl=0.0009
+update=8500 collisions=18.18 chaser_return=19.55 explorer_return=23.19 vision=(0.937,0.937) new_fields=(50.9,58.9) distance=2.01 loss=(-0.004,5.536) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_009000.safetensors
+update=9000 collisions=19.70 chaser_return=20.76 explorer_return=23.81 vision=(0.896,0.896) new_fields=(49.4,59.3) distance=2.05 loss=(-0.006,6.206) kl=0.0010
+update=9500 collisions=20.40 chaser_return=19.84 explorer_return=4.94 vision=(0.936,0.936) new_fields=(39.7,46.7) distance=1.39 loss=(-0.003,5.978) kl=0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_010000.safetensors
+update=10000 collisions=3.60 chaser_return=-0.96 explorer_return=52.08 vision=(0.477,0.477) new_fields=(26.4,70.8) distance=5.93 loss=(-0.002,6.723) kl=-0.0005
+update=10500 collisions=16.50 chaser_return=17.14 explorer_return=27.18 vision=(0.912,0.912) new_fields=(47.9,60.6) distance=1.75 loss=(-0.003,5.247) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_011000.safetensors
+update=11000 collisions=19.65 chaser_return=20.23 explorer_return=17.12 vision=(0.940,0.940) new_fields=(47.0,55.2) distance=1.72 loss=(-0.006,5.099) kl=0.0014
+update=11500 collisions=18.83 chaser_return=20.05 explorer_return=30.06 vision=(0.955,0.955) new_fields=(51.0,63.2) distance=1.88 loss=(-0.004,4.981) kl=0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_012000.safetensors
+update=12000 collisions=18.08 chaser_return=18.65 explorer_return=22.34 vision=(0.907,0.907) new_fields=(47.4,58.2) distance=1.80 loss=(-0.003,4.145) kl=0.0004
+update=12500 collisions=19.98 chaser_return=20.64 explorer_return=24.54 vision=(0.942,0.942) new_fields=(47.7,61.1) distance=1.73 loss=(-0.004,4.642) kl=0.0010
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_013000.safetensors
+update=13000 collisions=19.62 chaser_return=20.90 explorer_return=21.83 vision=(0.910,0.910) new_fields=(49.8,58.3) distance=1.62 loss=(-0.004,5.183) kl=-0.0001
+update=13500 collisions=26.38 chaser_return=27.62 explorer_return=16.20 vision=(0.914,0.914) new_fields=(47.5,57.1) distance=1.58 loss=(-0.004,5.743) kl=-0.0000
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_014000.safetensors
+update=14000 collisions=18.58 chaser_return=20.01 explorer_return=24.71 vision=(0.923,0.923) new_fields=(51.2,59.9) distance=1.70 loss=(-0.005,4.892) kl=0.0010
+update=14500 collisions=21.05 chaser_return=22.18 explorer_return=27.41 vision=(0.911,0.911) new_fields=(48.8,63.1) distance=2.06 loss=(-0.004,4.936) kl=0.0005
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_015000.safetensors
+update=15000 collisions=15.12 chaser_return=15.53 explorer_return=25.88 vision=(0.941,0.941) new_fields=(47.5,59.5) distance=1.76 loss=(-0.005,5.138) kl=0.0009
+update=15500 collisions=20.98 chaser_return=21.89 explorer_return=22.80 vision=(0.932,0.932) new_fields=(47.0,61.3) distance=1.71 loss=(-0.004,4.877) kl=0.0007
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_016000.safetensors
+update=16000 collisions=17.68 chaser_return=18.88 explorer_return=18.34 vision=(0.932,0.932) new_fields=(49.8,55.5) distance=1.68 loss=(-0.004,4.763) kl=0.0012
+update=16500 collisions=21.10 chaser_return=22.72 explorer_return=19.21 vision=(0.935,0.935) new_fields=(51.7,58.7) distance=1.96 loss=(-0.004,4.645) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_017000.safetensors
+update=17000 collisions=18.73 chaser_return=20.13 explorer_return=22.95 vision=(0.930,0.930) new_fields=(50.9,59.8) distance=1.79 loss=(-0.003,4.765) kl=0.0005
+update=17500 collisions=18.80 chaser_return=20.43 explorer_return=26.21 vision=(0.888,0.888) new_fields=(52.6,61.2) distance=1.85 loss=(-0.003,4.952) kl=0.0014
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_018000.safetensors
+update=18000 collisions=18.33 chaser_return=19.10 explorer_return=23.00 vision=(0.924,0.924) new_fields=(48.0,58.7) distance=1.95 loss=(-0.005,5.410) kl=0.0014
+update=18500 collisions=19.33 chaser_return=20.35 explorer_return=16.84 vision=(0.943,0.943) new_fields=(49.3,55.8) distance=1.61 loss=(-0.003,4.171) kl=0.0009
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/update_019000.safetensors
+update=19000 collisions=18.02 chaser_return=18.94 explorer_return=22.70 vision=(0.940,0.940) new_fields=(49.8,58.8) distance=1.75 loss=(-0.005,5.021) kl=0.0007
+update=19500 collisions=18.23 chaser_return=19.14 explorer_return=21.96 vision=(0.937,0.937) new_fields=(50.2,57.9) distance=1.57 loss=(-0.002,4.085) kl=0.0001
+update=20000 collisions=21.25 chaser_return=22.77 explorer_return=21.69 vision=(0.955,0.955) new_fields=(51.6,58.4) distance=1.54 loss=(-0.003,4.631) kl=-0.0003
+saved_checkpoint=runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors
+MRR_JOB_RESULT={"attempt_id": "attempt_01", "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors", "elapsed_seconds": 7585.706820899039, "metrics": {"approx_kl": -0.0002899035942244277, "chaser_grad_norm": 0.2534610331058502, "chaser_new_fields": 51.625, "chaser_partner_vision": 0.9552500247955322, "chaser_return": 22.76500129699707, "chaser_subspace_norm": 57.69636535644531, "collisions_per_episode": 21.25, "entropy": 0.1398237645626068, "explorer_grad_norm": 0.3390047997236252, "explorer_new_fields": 58.42499923706055, "explorer_partner_vision": 0.9552500247955322, "explorer_return": 21.6875, "explorer_subspace_norm": 105.33492279052734, "final_distance": 1.5442389249801636, "max_param_abs": 1.1976157426834106, "policy_loss": -0.002912267684005876, "value_loss": 4.631349444389343}, "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01", "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01", "seed": 9, "task": "social", "unit_id": "social/seed_0009"}
+=== exited 2026-07-02T11:50:30.964111+00:00 returncode=0 ===

manifest.json

@@ -0,0 +1,423 @@
+{
+  "config": {
+    "architecture": "mlp",
+    "batch_size": 40,
+    "checkpoint_every": 1000,
+    "checkpoint_every_seconds": 1800.0,
+    "clip_epsilon": 0.2,
+    "concurrency": 10,
+    "cuda_tf32": true,
+    "device": "cuda",
+    "entropy_coef": 0.01,
+    "finite_guard": true,
+    "fused_agent_rollout": false,
+    "grad_clip": 1.0,
+    "hidden_size": 256,
+    "learning_rate": 0.0003,
+    "max_steps": 100,
+    "ppo_epochs": 4,
+    "preset": "paper_text",
+    "recurrent_l2_coef": 0.3,
+    "status_every_seconds": 1800.0,
+    "subspace_metric_period": 10,
+    "triton_env_step": true,
+    "updates": 20000,
+    "value_clip": 10.0,
+    "value_coef": 0.5
+  },
+  "config_sha256": "1e164a3dab3a7ecde08c0595611a60ed85489e06861f477e46efeb1be25d94c3",
+  "created_at": "2026-07-02T07:36:16.447422+00:00",
+  "experiment": "mouse-run-run-2-mlp",
+  "experiment_spec": "experiments/paper_marl_2026/SPEC.md",
+  "experiment_spec_sha256": "e878bd03a06735c93c8caab4fc769ce63f026a538dfccbc87b588d432e48761e",
+  "initial_attempts": [
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0000_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0000/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0000.attempt_01",
+      "seed": 0,
+      "task": "social",
+      "unit_id": "social/seed_0000"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0000_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01",
+      "seed": 0,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0000"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0001_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0001/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0001.attempt_01",
+      "seed": 1,
+      "task": "social",
+      "unit_id": "social/seed_0001"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0001_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01",
+      "seed": 1,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0001"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0002_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0002/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0002.attempt_01",
+      "seed": 2,
+      "task": "social",
+      "unit_id": "social/seed_0002"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0002_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0002/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0002.attempt_01",
+      "seed": 2,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0002"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0003_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0003/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0003.attempt_01",
+      "seed": 3,
+      "task": "social",
+      "unit_id": "social/seed_0003"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0003_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0003/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0003.attempt_01",
+      "seed": 3,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0003"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0004_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0004/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0004.attempt_01",
+      "seed": 4,
+      "task": "social",
+      "unit_id": "social/seed_0004"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0004_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0004/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0004.attempt_01",
+      "seed": 4,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0004"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0005_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0005/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0005.attempt_01",
+      "seed": 5,
+      "task": "social",
+      "unit_id": "social/seed_0005"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0005_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0005/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0005.attempt_01",
+      "seed": 5,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0005"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0006_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0006/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0006.attempt_01",
+      "seed": 6,
+      "task": "social",
+      "unit_id": "social/seed_0006"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0006_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0006/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0006.attempt_01",
+      "seed": 6,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0006"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0007_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0007/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0007.attempt_01",
+      "seed": 7,
+      "task": "social",
+      "unit_id": "social/seed_0007"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0007_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0007/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0007.attempt_01",
+      "seed": 7,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0007"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0008_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0008/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0008.attempt_01",
+      "seed": 8,
+      "task": "social",
+      "unit_id": "social/seed_0008"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0008_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0008/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0008.attempt_01",
+      "seed": 8,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0008"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/social_seed_0009_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/social/seed_0009/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.social.seed_0009.attempt_01",
+      "seed": 9,
+      "task": "social",
+      "unit_id": "social/seed_0009"
+    },
+    {
+      "attempt": 1,
+      "attempt_id": "attempt_01",
+      "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01/checkpoints/latest.safetensors",
+      "log": "runs/mouse-run-run-2-mlp/logs/non_social_seed_0009_attempt_01.log",
+      "resume_from": null,
+      "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0009/attempt_01",
+      "run_id": "mouse-run-run-2-mlp.non_social.seed_0009.attempt_01",
+      "seed": 9,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0009"
+    }
+  ],
+  "max_attempts": 2,
+  "provenance": {
+    "command": "scripts/run_local_experiment.py --config experiments/paper_marl_2026/configs/arch_mlp.json",
+    "created_at": "2026-07-02T07:36:16.447522+00:00",
+    "dependencies": {
+      "huggingface_hub": "1.21.0",
+      "numpy": "2.5.0",
+      "safetensors": "0.8.0",
+      "tensorboard": "2.21.0",
+      "torch": "2.12.1",
+      "triton": "3.7.1"
+    },
+    "executable": "/root/mouse-run-run-perf/.venv/bin/python3",
+    "git": {
+      "branch": "main",
+      "diff_sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
+      "dirty": false,
+      "sha": "6282c650d281b31d646d537dd3f57c968806a423",
+      "short_sha": "6282c65",
+      "status_short": ""
+    },
+    "platform": "Linux-6.8.0-117-generic-x86_64-with-glibc2.39",
+    "python": "3.14.5 (main, Jun  2 2026, 22:25:17) [Clang 22.1.3 ]",
+    "schema_version": 1,
+    "torch": {
+      "cuda_available": true,
+      "cuda_device_count": 1,
+      "cuda_devices": [
+        {
+          "index": 0,
+          "major": 8,
+          "minor": 9,
+          "multi_processor_count": 66,
+          "name": "NVIDIA GeForce RTX 4070 Ti SUPER",
+          "total_memory_bytes": 16717578240
+        }
+      ],
+      "cuda_version": "13.0",
+      "cudnn_version": 92000,
+      "mps_available": false,
+      "version": "2.12.1+cu130"
+    }
+  },
+  "schema_version": 2,
+  "total_environment_steps": 1600000000,
+  "total_units": 20,
+  "units": [
+    {
+      "seed": 0,
+      "task": "social",
+      "unit_id": "social/seed_0000"
+    },
+    {
+      "seed": 0,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0000"
+    },
+    {
+      "seed": 1,
+      "task": "social",
+      "unit_id": "social/seed_0001"
+    },
+    {
+      "seed": 1,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0001"
+    },
+    {
+      "seed": 2,
+      "task": "social",
+      "unit_id": "social/seed_0002"
+    },
+    {
+      "seed": 2,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0002"
+    },
+    {
+      "seed": 3,
+      "task": "social",
+      "unit_id": "social/seed_0003"
+    },
+    {
+      "seed": 3,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0003"
+    },
+    {
+      "seed": 4,
+      "task": "social",
+      "unit_id": "social/seed_0004"
+    },
+    {
+      "seed": 4,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0004"
+    },
+    {
+      "seed": 5,
+      "task": "social",
+      "unit_id": "social/seed_0005"
+    },
+    {
+      "seed": 5,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0005"
+    },
+    {
+      "seed": 6,
+      "task": "social",
+      "unit_id": "social/seed_0006"
+    },
+    {
+      "seed": 6,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0006"
+    },
+    {
+      "seed": 7,
+      "task": "social",
+      "unit_id": "social/seed_0007"
+    },
+    {
+      "seed": 7,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0007"
+    },
+    {
+      "seed": 8,
+      "task": "social",
+      "unit_id": "social/seed_0008"
+    },
+    {
+      "seed": 8,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0008"
+    },
+    {
+      "seed": 9,
+      "task": "social",
+      "unit_id": "social/seed_0009"
+    },
+    {
+      "seed": 9,
+      "task": "non_social",
+      "unit_id": "non_social/seed_0009"
+    }
+  ]
+}

non_social/seed_0000/attempt_01/checkpoints/latest.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8e5409f712499d04d8944b4ecd7070e6a1688c886ed4463cf04bf85fc9707a3
+size 11460376

non_social/seed_0000/attempt_01/config.json

@@ -0,0 +1,44 @@
+{
+  "architecture": "mlp",
+  "attempt_id": "attempt_01",
+  "batch_size": 40,
+  "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors",
+  "checkpoint_every": 1000,
+  "checkpoint_every_seconds": 1800.0,
+  "clip_epsilon": 0.2,
+  "cost_per_hour": null,
+  "cuda_tf32": true,
+  "device": "cuda",
+  "entropy_coef": 0.01,
+  "env": {
+    "grid_size": 10,
+    "max_steps": 100,
+    "partner_visibility": "none",
+    "task": "non_social",
+    "vision_radius": 3
+  },
+  "experiment_id": "mouse-run-run-2-mlp",
+  "finite_guard": true,
+  "fused_agent_rollout": false,
+  "gae_lambda": 0.95,
+  "gamma": 0.99,
+  "grad_clip": 1.0,
+  "hidden_size": 256,
+  "learning_rate": 0.0003,
+  "log_every": 500,
+  "metrics_path": null,
+  "ppo_epochs": 4,
+  "recurrent_l2_coef": 0.3,
+  "resume_from": null,
+  "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01",
+  "run_id": "mouse-run-run-2-mlp.non_social.seed_0000.attempt_01",
+  "seed": 0,
+  "status_every_seconds": 1800.0,
+  "status_path": null,
+  "subspace_metric_period": 10,
+  "tensorboard_dir": null,
+  "triton_env_step": true,
+  "updates": 20000,
+  "value_clip": 10.0,
+  "value_coef": 0.5
+}

non_social/seed_0000/attempt_01/metrics.jsonl

@@ -0,0 +1,42 @@
+{"device": "cuda", "elapsed_hours": 0.0005362351288850833, "elapsed_seconds": 1.9304464639863, "env_steps_per_second": 2072.059533699862, "environment_steps": 4000, "error": null, "estimated_cost_usd": null, "eta_hours": 10.72416634257278, "eta_seconds": 38606.998833262005, "healthy": true, "latest_checkpoint": null, "metrics": {"approx_kl": -5.173608782563233e-06, "chaser_grad_norm": 0.34016749262809753, "chaser_new_fields": 35.775001525878906, "chaser_partner_vision": 0.0, "chaser_return": -2.9100003242492676, "chaser_subspace_norm": 0.0029390037525445223, "collisions_per_episode": 1.7000000476837158, "entropy": 2.7725364565849304, "explorer_grad_norm": 0.1394939087331295, "explorer_new_fields": 33.70000076293945, "explorer_partner_vision": 0.0, "explorer_return": 0.10000000149011612, "explorer_subspace_norm": 0.0024945889599621296, "final_distance": 5.466697692871094, "max_param_abs": 0.15428763628005981, "policy_loss": -0.000497843058607117, "value_loss": 3.074083089828491}, "progress_fraction": 5e-05, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:36:19.653523+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782977779.6535435, "update": 1}
+{"device": "cuda", "elapsed_hours": 0.05329365353998987, "elapsed_seconds": 191.85715274396352, "env_steps_per_second": 10424.422396537033, "environment_steps": 2000000, "error": null, "estimated_cost_usd": null, "eta_hours": 2.078452488059605, "eta_seconds": 7482.428957014578, "healthy": true, "latest_checkpoint": null, "metrics": {"approx_kl": 6.691479853282445e-05, "chaser_grad_norm": 0.23669546842575073, "chaser_new_fields": 74.20000457763672, "chaser_partner_vision": 0.0, "chaser_return": 4.700000762939453, "chaser_subspace_norm": 16.239971160888672, "collisions_per_episode": 3.375, "entropy": 0.4663320481777191, "explorer_grad_norm": 0.3577277585864067, "explorer_new_fields": 67.32500457763672, "explorer_partner_vision": 0.0, "explorer_return": 49.9375, "explorer_subspace_norm": 8.448774337768555, "final_distance": 4.963988780975342, "max_param_abs": 0.3258746862411499, "policy_loss": -0.0018085034082275797, "value_loss": 4.557863235473633}, "progress_fraction": 0.025, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:39:29.580229+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782977969.5802422, "update": 500}
+{"device": "cuda", "elapsed_hours": 0.10710371320084151, "elapsed_seconds": 385.57336752302945, "env_steps_per_second": 10374.160502050467, "environment_steps": 4000000, "error": null, "estimated_cost_usd": null, "eta_hours": 2.0349705508159883, "eta_seconds": 7325.893982937559, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_001000.safetensors", "metrics": {"approx_kl": 0.0007281143502567144, "chaser_grad_norm": 0.17577309906482697, "chaser_new_fields": 79.5, "chaser_partner_vision": 0.0, "chaser_return": 5.810000896453857, "chaser_subspace_norm": 26.5023136138916, "collisions_per_episode": 1.899999976158142, "entropy": 0.3493778705596924, "explorer_grad_norm": 0.6742401272058487, "explorer_new_fields": 67.3499984741211, "explorer_partner_vision": 0.0, "explorer_return": 50.42499923706055, "explorer_subspace_norm": 8.997306823730469, "final_distance": 5.553553581237793, "max_param_abs": 0.3305240869522095, "policy_loss": -0.0016281890529228349, "value_loss": 4.399340271949768}, "progress_fraction": 0.05, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:42:43.296439+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978163.2964523, "update": 1000}
+{"device": "cuda", "elapsed_hours": 0.1595268888483406, "elapsed_seconds": 574.2967998540262, "env_steps_per_second": 10447.559522402127, "environment_steps": 6000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.9674982957962008, "eta_seconds": 7082.993864866323, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_001000.safetensors", "metrics": {"approx_kl": 0.0006018214631288155, "chaser_grad_norm": 0.21672200597822666, "chaser_new_fields": 80.8499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.075000762939453, "chaser_subspace_norm": 36.892730712890625, "collisions_per_episode": 3.049999952316284, "entropy": 0.3051275759935379, "explorer_grad_norm": 0.4998535215854645, "explorer_new_fields": 69.5999984741211, "explorer_partner_vision": 0.0, "explorer_return": 53.650001525878906, "explorer_subspace_norm": 8.920053482055664, "final_distance": 5.113457679748535, "max_param_abs": 0.349557101726532, "policy_loss": -0.0010127461082305445, "value_loss": 3.9467450380325317}, "progress_fraction": 0.075, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:45:52.019871+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978352.0198798, "update": 1500}
+{"device": "cuda", "elapsed_hours": 0.21280761047722382, "elapsed_seconds": 766.1073977180058, "env_steps_per_second": 10442.400143673714, "environment_steps": 8000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.915268494295015, "eta_seconds": 6894.966579462053, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_002000.safetensors", "metrics": {"approx_kl": 0.0008088869087739536, "chaser_grad_norm": 0.07538860198110342, "chaser_new_fields": 79.6500015258789, "chaser_partner_vision": 0.0, "chaser_return": 5.820001125335693, "chaser_subspace_norm": 42.316566467285156, "collisions_per_episode": 2.125, "entropy": 0.30866675823926926, "explorer_grad_norm": 1.8579207435250282, "explorer_new_fields": 67.32500457763672, "explorer_partner_vision": 0.0, "explorer_return": 50.42499923706055, "explorer_subspace_norm": 10.101943969726562, "final_distance": 4.400704383850098, "max_param_abs": 0.4545920789241791, "policy_loss": -0.0013379232555177367, "value_loss": 4.046706676483154}, "progress_fraction": 0.1, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:49:03.830470+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978543.8304827, "update": 2000}
+{"device": "cuda", "elapsed_hours": 0.2660434453658268, "elapsed_seconds": 957.7564033169765, "env_steps_per_second": 10441.068277243798, "environment_steps": 10000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.8623041175607877, "eta_seconds": 6704.2948232188355, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_002000.safetensors", "metrics": {"approx_kl": 0.0006254628758133318, "chaser_grad_norm": 0.2807302428409457, "chaser_new_fields": 76.70000457763672, "chaser_partner_vision": 0.0, "chaser_return": 5.1800007820129395, "chaser_subspace_norm": 51.4793701171875, "collisions_per_episode": 2.299999952316284, "entropy": 0.23320892453193665, "explorer_grad_norm": 1.430855542421341, "explorer_new_fields": 70.82500457763672, "explorer_partner_vision": 0.0, "explorer_return": 55.11249923706055, "explorer_subspace_norm": 10.984577178955078, "final_distance": 5.455503940582275, "max_param_abs": 0.511533260345459, "policy_loss": -0.0011108804219404078, "value_loss": 3.958456575870514}, "progress_fraction": 0.125, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:52:15.479474+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978735.4794846, "update": 2500}
+{"device": "cuda", "elapsed_hours": 0.3191621872458361, "elapsed_seconds": 1148.98387408501, "env_steps_per_second": 10444.010808729727, "environment_steps": 12000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.8085857277264048, "eta_seconds": 6510.908619815057, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_003000.safetensors", "metrics": {"approx_kl": 0.00035122274371035545, "chaser_grad_norm": 0.10724705085158348, "chaser_new_fields": 77.55000305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.355000972747803, "chaser_subspace_norm": 54.21003341674805, "collisions_per_episode": 2.375, "entropy": 0.26491987705230713, "explorer_grad_norm": 0.5290967896580696, "explorer_new_fields": 67.55000305175781, "explorer_partner_vision": 0.0, "explorer_return": 50.57500076293945, "explorer_subspace_norm": 10.944963455200195, "final_distance": 5.601850509643555, "max_param_abs": 0.5453172922134399, "policy_loss": -0.0015410876681567487, "value_loss": 3.992495119571686}, "progress_fraction": 0.15, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:55:26.706946+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978926.7069602, "update": 3000}
+{"device": "cuda", "elapsed_hours": 0.37103237310637877, "elapsed_seconds": 1335.7165431829635, "env_steps_per_second": 10481.265708245639, "environment_steps": 14000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.7491526160729283, "eta_seconds": 6296.949417862542, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_003000.safetensors", "metrics": {"approx_kl": 0.0001249361482511535, "chaser_grad_norm": 0.12887633964419365, "chaser_new_fields": 80.70000457763672, "chaser_partner_vision": 0.0, "chaser_return": 6.02000093460083, "chaser_subspace_norm": 57.67909240722656, "collisions_per_episode": 1.899999976158142, "entropy": 0.2018074169754982, "explorer_grad_norm": 0.512401819229126, "explorer_new_fields": 70.0250015258789, "explorer_partner_vision": 0.0, "explorer_return": 54.36249923706055, "explorer_subspace_norm": 11.544838905334473, "final_distance": 4.9441657066345215, "max_param_abs": 0.5763495564460754, "policy_loss": -0.00044214877009096654, "value_loss": 3.771480619907379}, "progress_fraction": 0.175, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:58:33.439614+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979113.4396262, "update": 3500}
+{"device": "cuda", "elapsed_hours": 0.4231054240669538, "elapsed_seconds": 1523.1795266410336, "env_steps_per_second": 10504.342869736265, "environment_steps": 16000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.6924216962678151, "eta_seconds": 6092.718106564134, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_004000.safetensors", "metrics": {"approx_kl": 0.00014633760644949811, "chaser_grad_norm": 0.054581169970333576, "chaser_new_fields": 79.80000305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.875001430511475, "chaser_subspace_norm": 62.2651252746582, "collisions_per_episode": 1.649999976158142, "entropy": 0.19518577307462692, "explorer_grad_norm": 0.6297692358493805, "explorer_new_fields": 71.875, "explorer_partner_vision": 0.0, "explorer_return": 57.17499923706055, "explorer_subspace_norm": 12.46562671661377, "final_distance": 5.8548688888549805, "max_param_abs": 0.5877197980880737, "policy_loss": -0.0008621873332899099, "value_loss": 3.66826194524765}, "progress_fraction": 0.2, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:01:40.902600+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979300.9026148, "update": 4000}
+{"device": "cuda", "elapsed_hours": 0.4754179403783443, "elapsed_seconds": 1711.5045853620395, "env_steps_per_second": 10517.062094924162, "environment_steps": 18000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.637550683525408, "eta_seconds": 5895.182460691469, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_004000.safetensors", "metrics": {"approx_kl": 0.000319335286664002, "chaser_grad_norm": 0.12068943493068218, "chaser_new_fields": 81.8499984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.285000801086426, "chaser_subspace_norm": 70.37218475341797, "collisions_per_episode": 1.625, "entropy": 0.179047342389822, "explorer_grad_norm": 0.524848110973835, "explorer_new_fields": 69.8499984741211, "explorer_partner_vision": 0.0, "explorer_return": 54.13750076293945, "explorer_subspace_norm": 13.666532516479492, "final_distance": 5.531441688537598, "max_param_abs": 0.5752404928207397, "policy_loss": -0.0008478180620574705, "value_loss": 3.7342492938041687}, "progress_fraction": 0.225, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:04:49.227656+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979489.227666, "update": 4500}
+{"device": "cuda", "elapsed_hours": 0.5281459557525038, "elapsed_seconds": 1901.3254407090135, "env_steps_per_second": 10518.977746672292, "environment_steps": 20000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.5844378672575112, "eta_seconds": 5703.9763221270405, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_005000.safetensors", "metrics": {"approx_kl": 0.0008280922713789496, "chaser_grad_norm": 0.11795545369386673, "chaser_new_fields": 77.42500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.380000591278076, "chaser_subspace_norm": 69.02265167236328, "collisions_per_episode": 2.1500000953674316, "entropy": 0.14308607205748558, "explorer_grad_norm": 0.48360154777765274, "explorer_new_fields": 73.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 59.087501525878906, "explorer_subspace_norm": 13.96146011352539, "final_distance": 5.5770416259765625, "max_param_abs": 0.647881269454956, "policy_loss": -0.00018160262168720465, "value_loss": 3.802269697189331}, "progress_fraction": 0.25, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:07:59.048512+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979679.048526, "update": 5000}
+{"device": "cuda", "elapsed_hours": 0.5800792278741654, "elapsed_seconds": 2088.2852203469956, "env_steps_per_second": 10534.959394265317, "environment_steps": 22000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.5292997825773451, "eta_seconds": 5505.479217278443, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_005000.safetensors", "metrics": {"approx_kl": -4.474588249175904e-05, "chaser_grad_norm": 0.13463049568235874, "chaser_new_fields": 78.75, "chaser_partner_vision": 0.0, "chaser_return": 5.690001010894775, "chaser_subspace_norm": 68.8963623046875, "collisions_per_episode": 1.625, "entropy": 0.13056070357561111, "explorer_grad_norm": 0.8507837206125259, "explorer_new_fields": 72.80000305175781, "explorer_partner_vision": 0.0, "explorer_return": 58.525001525878906, "explorer_subspace_norm": 15.163471221923828, "final_distance": 5.202518463134766, "max_param_abs": 0.7021549940109253, "policy_loss": -0.0006455526853557103, "value_loss": 3.8411065340042114}, "progress_fraction": 0.275, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:11:06.008292+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979866.008303, "update": 5500}
+{"device": "cuda", "elapsed_hours": 0.6321262312230667, "elapsed_seconds": 2275.65443240304, "env_steps_per_second": 10546.416739845925, "environment_steps": 24000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.4749612061871558, "eta_seconds": 5309.860342273761, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_006000.safetensors", "metrics": {"approx_kl": 0.00040526999702991673, "chaser_grad_norm": 0.08878211490809917, "chaser_new_fields": 78.3499984741211, "chaser_partner_vision": 0.0, "chaser_return": 5.560000896453857, "chaser_subspace_norm": 68.70912170410156, "collisions_per_episode": 1.850000023841858, "entropy": 0.13891512528061867, "explorer_grad_norm": 0.6218316107988358, "explorer_new_fields": 72.42500305175781, "explorer_partner_vision": 0.0, "explorer_return": 57.70000076293945, "explorer_subspace_norm": 16.02243423461914, "final_distance": 4.691096305847168, "max_param_abs": 0.7957910895347595, "policy_loss": -0.0008608128022196126, "value_loss": 3.6975609064102173}, "progress_fraction": 0.3, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:14:13.377505+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980053.3775182, "update": 6000}
+{"device": "cuda", "elapsed_hours": 0.6850991590336263, "elapsed_seconds": 2466.3569725210546, "env_steps_per_second": 10541.864089294173, "environment_steps": 26000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.4228982533775314, "eta_seconds": 5122.433712159113, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_006000.safetensors", "metrics": {"approx_kl": 0.0002404470562145239, "chaser_grad_norm": 0.09046696685254574, "chaser_new_fields": 82.5999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.440001010894775, "chaser_subspace_norm": 77.56379699707031, "collisions_per_episode": 1.725000023841858, "entropy": 0.11232517659664154, "explorer_grad_norm": 0.6239490285515785, "explorer_new_fields": 72.92500305175781, "explorer_partner_vision": 0.0, "explorer_return": 58.60000228881836, "explorer_subspace_norm": 17.372209548950195, "final_distance": 5.052657604217529, "max_param_abs": 0.849467933177948, "policy_loss": -0.0007943277134287996, "value_loss": 3.6960580348968506}, "progress_fraction": 0.325, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:17:24.080043+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980244.080054, "update": 6500}
+{"device": "cuda", "elapsed_hours": 0.7369473368947123, "elapsed_seconds": 2653.010412820964, "env_steps_per_second": 10554.048285934698, "environment_steps": 28000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.3686164828044656, "eta_seconds": 4927.019338096076, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_007000.safetensors", "metrics": {"approx_kl": -1.6540823666894022e-05, "chaser_grad_norm": 0.11463696509599686, "chaser_new_fields": 82.55000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.4250006675720215, "chaser_subspace_norm": 78.05644226074219, "collisions_per_episode": 1.600000023841858, "entropy": 0.11654260009527206, "explorer_grad_norm": 0.8265438824892044, "explorer_new_fields": 77.45000457763672, "explorer_partner_vision": 0.0, "explorer_return": 65.5, "explorer_subspace_norm": 16.70997428894043, "final_distance": 4.866159915924072, "max_param_abs": 0.862048327922821, "policy_loss": -0.0003952661583070949, "value_loss": 3.6717289686203003}, "progress_fraction": 0.35, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:20:30.733484+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980430.733498, "update": 7000}
+{"device": "cuda", "elapsed_hours": 0.7888488838738865, "elapsed_seconds": 2839.8559819459915, "env_steps_per_second": 10563.915984022087, "environment_steps": 30000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.3147481397898109, "eta_seconds": 4733.0933032433195, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_007000.safetensors", "metrics": {"approx_kl": 0.00047657695344544004, "chaser_grad_norm": 0.06467235274612904, "chaser_new_fields": 80.25, "chaser_partner_vision": 0.0, "chaser_return": 5.950000762939453, "chaser_subspace_norm": 79.99512481689453, "collisions_per_episode": 1.774999976158142, "entropy": 0.09236839413642883, "explorer_grad_norm": 1.273687481880188, "explorer_new_fields": 77.25, "explorer_partner_vision": 0.0, "explorer_return": 65.05000305175781, "explorer_subspace_norm": 18.648405075073242, "final_distance": 5.496772766113281, "max_param_abs": 0.8750737905502319, "policy_loss": -0.0006252801441561928, "value_loss": 3.7000213265419006}, "progress_fraction": 0.375, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:23:37.579052+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980617.579064, "update": 7500}
+{"device": "cuda", "elapsed_hours": 0.840673852563066, "elapsed_seconds": 3026.4258692270378, "env_steps_per_second": 10573.528440058219, "environment_steps": 32000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.261010778844599, "eta_seconds": 4539.638803840557, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_008000.safetensors", "metrics": {"approx_kl": 1.1703377388594394e-06, "chaser_grad_norm": 0.20380252599716187, "chaser_new_fields": 80.30000305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.945001125335693, "chaser_subspace_norm": 82.45390319824219, "collisions_per_episode": 1.649999976158142, "entropy": 0.10570455901324749, "explorer_grad_norm": 1.6526433825492859, "explorer_new_fields": 75.9749984741211, "explorer_partner_vision": 0.0, "explorer_return": 63.25, "explorer_subspace_norm": 18.309865951538086, "final_distance": 5.806778430938721, "max_param_abs": 0.9599581360816956, "policy_loss": -0.0004156225741596131, "value_loss": 3.573492467403412}, "progress_fraction": 0.4, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:26:44.148942+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980804.1489553, "update": 8000}
+{"device": "cuda", "elapsed_hours": 0.8938823972936047, "elapsed_seconds": 3217.976630256977, "env_steps_per_second": 10565.645405972036, "environment_steps": 34000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.2093703022207591, "eta_seconds": 4353.733087994733, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_008000.safetensors", "metrics": {"approx_kl": 0.0004711518696320638, "chaser_grad_norm": 0.043468572199344635, "chaser_new_fields": 81.4000015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.185000896453857, "chaser_subspace_norm": 75.3616714477539, "collisions_per_episode": 1.9250000715255737, "entropy": 0.11739776842296124, "explorer_grad_norm": 0.8637182414531708, "explorer_new_fields": 76.67500305175781, "explorer_partner_vision": 0.0, "explorer_return": 64.5625, "explorer_subspace_norm": 19.596525192260742, "final_distance": 5.308167934417725, "max_param_abs": 1.043190836906433, "policy_loss": -0.0012657396960946699, "value_loss": 3.582657754421234}, "progress_fraction": 0.425, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:29:55.699701+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980995.6997137, "update": 8500}
+{"device": "cuda", "elapsed_hours": 0.9460653386549933, "elapsed_seconds": 3405.835219157976, "env_steps_per_second": 10570.094465374714, "environment_steps": 36000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.1563020805783253, "eta_seconds": 4162.687490081971, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_009000.safetensors", "metrics": {"approx_kl": 0.0004890187930004786, "chaser_grad_norm": 0.12140505574643612, "chaser_new_fields": 79.30000305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.7400007247924805, "chaser_subspace_norm": 82.12152099609375, "collisions_per_episode": 1.8250000476837158, "entropy": 0.10391798242926598, "explorer_grad_norm": 1.8681467473506927, "explorer_new_fields": 78.0250015258789, "explorer_partner_vision": 0.0, "explorer_return": 66.32500457763672, "explorer_subspace_norm": 18.82545280456543, "final_distance": 4.985391139984131, "max_param_abs": 1.093315839767456, "policy_loss": -0.0009176402622799174, "value_loss": 3.6649279594421387}, "progress_fraction": 0.45, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:33:03.558290+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981183.5583043, "update": 9000}
+{"device": "cuda", "elapsed_hours": 0.9989255789939003, "elapsed_seconds": 3596.1320843780413, "env_steps_per_second": 10566.908864409017, "environment_steps": 38000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.1040756399406266, "eta_seconds": 3974.672303786256, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_009000.safetensors", "metrics": {"approx_kl": 0.00019615038952985886, "chaser_grad_norm": 0.14012088812887669, "chaser_new_fields": 81.30000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.160000801086426, "chaser_subspace_norm": 80.92264556884766, "collisions_per_episode": 1.8250000476837158, "entropy": 0.09415227919816971, "explorer_grad_norm": 0.38225405663251877, "explorer_new_fields": 77.0999984741211, "explorer_partner_vision": 0.0, "explorer_return": 64.9375, "explorer_subspace_norm": 20.194814682006836, "final_distance": 4.540725231170654, "max_param_abs": 1.180087924003601, "policy_loss": -0.000374201442522093, "value_loss": 3.4838927388191223}, "progress_fraction": 0.475, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:36:13.855155+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981373.8551664, "update": 9500}
+{"device": "cuda", "elapsed_hours": 1.0525457070244333, "elapsed_seconds": 3789.16454528796, "env_steps_per_second": 10556.4167303693, "environment_steps": 40000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.0525457070244333, "eta_seconds": 3789.16454528796, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_010000.safetensors", "metrics": {"approx_kl": -0.0001693130862814201, "chaser_grad_norm": 0.07559763826429844, "chaser_new_fields": 81.4000015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.155000686645508, "chaser_subspace_norm": 85.33003997802734, "collisions_per_episode": 2.25, "entropy": 0.08448796719312668, "explorer_grad_norm": 1.4887766242027283, "explorer_new_fields": 77.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 65.2750015258789, "explorer_subspace_norm": 21.593774795532227, "final_distance": 5.322676181793213, "max_param_abs": 1.2174458503723145, "policy_loss": -1.3907546148272942e-05, "value_loss": 3.5226200222969055}, "progress_fraction": 0.5, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:39:26.887618+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981566.8876326, "update": 10000}
+{"device": "cuda", "elapsed_hours": 1.1047508845430436, "elapsed_seconds": 3977.103184354957, "env_steps_per_second": 10560.450170168753, "environment_steps": 42000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.9995365145865633, "eta_seconds": 3598.331452511628, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_010000.safetensors", "metrics": {"approx_kl": -1.5043517249679894e-06, "chaser_grad_norm": 0.05930179636925459, "chaser_new_fields": 82.67500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.415000915527344, "chaser_subspace_norm": 86.52581787109375, "collisions_per_episode": 2.0, "entropy": 0.06573069468140602, "explorer_grad_norm": 0.4400622770190239, "explorer_new_fields": 77.07500457763672, "explorer_partner_vision": 0.0, "explorer_return": 65.20000457763672, "explorer_subspace_norm": 22.82740592956543, "final_distance": 5.1441144943237305, "max_param_abs": 1.2872856855392456, "policy_loss": -0.0009458466398069021, "value_loss": 3.4636882543563843}, "progress_fraction": 0.525, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:42:34.826255+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981754.8262658, "update": 10500}
+{"device": "cuda", "elapsed_hours": 1.158430229554716, "elapsed_seconds": 4170.3488263969775, "env_steps_per_second": 10550.676174015478, "environment_steps": 44000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.9478065514538585, "eta_seconds": 3412.1035852338905, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_011000.safetensors", "metrics": {"approx_kl": 7.785861745286671e-05, "chaser_grad_norm": 0.04884711094200611, "chaser_new_fields": 83.07500457763672, "chaser_partner_vision": 0.0, "chaser_return": 6.520001411437988, "chaser_subspace_norm": 87.99630737304688, "collisions_per_episode": 1.524999976158142, "entropy": 0.06133109796792269, "explorer_grad_norm": 0.802131175994873, "explorer_new_fields": 78.25, "explorer_partner_vision": 0.0, "explorer_return": 66.5875015258789, "explorer_subspace_norm": 23.026655197143555, "final_distance": 4.931851863861084, "max_param_abs": 1.2872856855392456, "policy_loss": -0.00041549857851563843, "value_loss": 3.494303584098816}, "progress_fraction": 0.55, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:45:48.071898+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981948.0719104, "update": 11000}
+{"device": "cuda", "elapsed_hours": 1.2102489595363861, "elapsed_seconds": 4356.89625433099, "env_steps_per_second": 10557.974602740085, "environment_steps": 46000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.8945318396573289, "eta_seconds": 3220.3146227663838, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_011000.safetensors", "metrics": {"approx_kl": 8.80808145280465e-05, "chaser_grad_norm": 0.07423081062734127, "chaser_new_fields": 80.80000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.070000648498535, "chaser_subspace_norm": 80.28471374511719, "collisions_per_episode": 1.399999976158142, "entropy": 0.061599936336278915, "explorer_grad_norm": 1.113692745566368, "explorer_new_fields": 79.95000457763672, "explorer_partner_vision": 0.0, "explorer_return": 69.32500457763672, "explorer_subspace_norm": 24.018014907836914, "final_distance": 5.806375503540039, "max_param_abs": 1.3213297128677368, "policy_loss": -0.0020681732161946798, "value_loss": 3.4564942121505737}, "progress_fraction": 0.575, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:48:54.619325+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982134.6193366, "update": 11500}
+{"device": "cuda", "elapsed_hours": 1.2624173470558404, "elapsed_seconds": 4544.702449401026, "env_steps_per_second": 10561.747558704577, "environment_steps": 48000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.8416115647038938, "eta_seconds": 3029.8016329340176, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_012000.safetensors", "metrics": {"approx_kl": 0.0008321205401692622, "chaser_grad_norm": 0.12845297157764435, "chaser_new_fields": 75.875, "chaser_partner_vision": 0.0, "chaser_return": 5.105000972747803, "chaser_subspace_norm": 81.34334564208984, "collisions_per_episode": 1.649999976158142, "entropy": 0.09196563623845577, "explorer_grad_norm": 0.29978178441524506, "explorer_new_fields": 77.9749984741211, "explorer_partner_vision": 0.0, "explorer_return": 66.17500305175781, "explorer_subspace_norm": 24.92875862121582, "final_distance": 5.172975063323975, "max_param_abs": 1.3213297128677368, "policy_loss": -0.0010934120214081133, "value_loss": 3.763103723526001}, "progress_fraction": 0.6, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:52:02.425521+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982322.425536, "update": 12000}
+{"device": "cuda", "elapsed_hours": 1.3171340958397357, "elapsed_seconds": 4741.6827450230485, "env_steps_per_second": 10544.779709794135, "environment_steps": 50000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.7902804575038415, "eta_seconds": 2845.009647013829, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_012000.safetensors", "metrics": {"approx_kl": 0.00044572072633515536, "chaser_grad_norm": 0.07732724957168102, "chaser_new_fields": 84.30000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.785000801086426, "chaser_subspace_norm": 78.33721160888672, "collisions_per_episode": 1.350000023841858, "entropy": 0.08976677805185318, "explorer_grad_norm": 1.0155585557222366, "explorer_new_fields": 79.125, "explorer_partner_vision": 0.0, "explorer_return": 68.125, "explorer_subspace_norm": 23.136762619018555, "final_distance": 5.725672245025635, "max_param_abs": 1.3213297128677368, "policy_loss": -0.001024966465664079, "value_loss": 3.352021098136902}, "progress_fraction": 0.625, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:55:19.405818+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982519.4058306, "update": 12500}
+{"device": "cuda", "elapsed_hours": 1.3712515359733435, "elapsed_seconds": 4936.5055295040365, "env_steps_per_second": 10533.767194062955, "environment_steps": 52000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.7383662116779541, "eta_seconds": 2658.118362040635, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_013000.safetensors", "metrics": {"approx_kl": 0.0004590160541175642, "chaser_grad_norm": 0.06538167875260115, "chaser_new_fields": 83.2249984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.54500150680542, "chaser_subspace_norm": 78.45780181884766, "collisions_per_episode": 1.399999976158142, "entropy": 0.09547889046370983, "explorer_grad_norm": 0.5435245260596275, "explorer_new_fields": 78.70000457763672, "explorer_partner_vision": 0.0, "explorer_return": 67.63750457763672, "explorer_subspace_norm": 24.118432998657227, "final_distance": 5.541812419891357, "max_param_abs": 1.3252073526382446, "policy_loss": -0.0007060517087804641, "value_loss": 3.4721049070358276}, "progress_fraction": 0.65, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:58:34.228603+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982714.2286181, "update": 13000}
+{"device": "cuda", "elapsed_hours": 1.4240544896291816, "elapsed_seconds": 5126.596162665053, "env_steps_per_second": 10533.304806268996, "environment_steps": 54000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.6856558653770133, "eta_seconds": 2468.361115357248, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_013000.safetensors", "metrics": {"approx_kl": 0.0008695930212665459, "chaser_grad_norm": 0.0758396191522479, "chaser_new_fields": 82.5999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.42000150680542, "chaser_subspace_norm": 74.20014190673828, "collisions_per_episode": 1.725000023841858, "entropy": 0.050692721270024776, "explorer_grad_norm": 1.3435713052749634, "explorer_new_fields": 80.32500457763672, "explorer_partner_vision": 0.0, "explorer_return": 69.70000457763672, "explorer_subspace_norm": 25.260629653930664, "final_distance": 5.570320129394531, "max_param_abs": 1.3920022249221802, "policy_loss": -0.0019914439972481546, "value_loss": 3.334362030029297}, "progress_fraction": 0.675, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:01:44.319234+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982904.3192446, "update": 13500}
+{"device": "cuda", "elapsed_hours": 1.4773909531022138, "elapsed_seconds": 5318.6074311679695, "env_steps_per_second": 10529.071890478363, "environment_steps": 56000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.6331675513295202, "eta_seconds": 2279.4031847862725, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_014000.safetensors", "metrics": {"approx_kl": 0.0002321924548440088, "chaser_grad_norm": 0.06448401790112257, "chaser_new_fields": 82.82500457763672, "chaser_partner_vision": 0.0, "chaser_return": 6.415000915527344, "chaser_subspace_norm": 81.34850311279297, "collisions_per_episode": 1.8250000476837158, "entropy": 0.06315584108233452, "explorer_grad_norm": 0.8479587733745575, "explorer_new_fields": 78.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 66.625, "explorer_subspace_norm": 27.18760871887207, "final_distance": 5.711249351501465, "max_param_abs": 1.4050794839859009, "policy_loss": -0.0011196303819325237, "value_loss": 3.3451501727104187}, "progress_fraction": 0.7, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:04:56.330503+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983096.330517, "update": 14000}
+{"device": "cuda", "elapsed_hours": 1.5296515264125061, "elapsed_seconds": 5506.745495085022, "env_steps_per_second": 10532.536877138626, "environment_steps": 58000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.5802126479495713, "eta_seconds": 2088.7655326184567, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_014000.safetensors", "metrics": {"approx_kl": -6.950403765165585e-05, "chaser_grad_norm": 0.062295411713421345, "chaser_new_fields": 82.55000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.390000820159912, "chaser_subspace_norm": 79.73140716552734, "collisions_per_episode": 1.7000000476837158, "entropy": 0.05496581457555294, "explorer_grad_norm": 1.555381953716278, "explorer_new_fields": 79.125, "explorer_partner_vision": 0.0, "explorer_return": 68.01250457763672, "explorer_subspace_norm": 26.886980056762695, "final_distance": 5.146921634674072, "max_param_abs": 1.4050794839859009, "policy_loss": -0.00048470002552480906, "value_loss": 3.3966342210769653}, "progress_fraction": 0.725, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:08:04.468566+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983284.4685764, "update": 14500}
+{"device": "cuda", "elapsed_hours": 1.5815020676722311, "elapsed_seconds": 5693.407443620032, "env_steps_per_second": 10538.50450616095, "environment_steps": 60000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.5271673558907437, "eta_seconds": 1897.8024812066772, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_015000.safetensors", "metrics": {"approx_kl": 0.00032535001055494783, "chaser_grad_norm": 0.0946476785466075, "chaser_new_fields": 84.4749984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.770000457763672, "chaser_subspace_norm": 81.79399871826172, "collisions_per_episode": 1.399999976158142, "entropy": 0.08795448578894138, "explorer_grad_norm": 1.0476092919707298, "explorer_new_fields": 78.7249984741211, "explorer_partner_vision": 0.0, "explorer_return": 67.5250015258789, "explorer_subspace_norm": 25.196977615356445, "final_distance": 5.499534606933594, "max_param_abs": 1.420541524887085, "policy_loss": -0.0003041045347949023, "value_loss": 3.2325546741485596}, "progress_fraction": 0.75, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:11:11.130515+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983471.1305299, "update": 15000}
+{"device": "cuda", "elapsed_hours": 1.6352512755986148, "elapsed_seconds": 5886.904592155013, "env_steps_per_second": 10531.850657580255, "environment_steps": 62000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.4747503703350817, "eta_seconds": 1709.1013332062942, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_015000.safetensors", "metrics": {"approx_kl": -8.603196387602097e-05, "chaser_grad_norm": 0.05303043685853481, "chaser_new_fields": 83.05000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.555000305175781, "chaser_subspace_norm": 80.69808197021484, "collisions_per_episode": 1.1750000715255737, "entropy": 0.08621848374605179, "explorer_grad_norm": 0.94834154099226, "explorer_new_fields": 76.2249984741211, "explorer_partner_vision": 0.0, "explorer_return": 63.85000228881836, "explorer_subspace_norm": 24.56145477294922, "final_distance": 5.3362135887146, "max_param_abs": 1.435640811920166, "policy_loss": -0.0006088202145186905, "value_loss": 3.4417847990989685}, "progress_fraction": 0.775, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:14:24.627665+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983664.6276793, "update": 15500}
+{"device": "cuda", "elapsed_hours": 1.6885142894677767, "elapsed_seconds": 6078.651442083996, "env_steps_per_second": 10528.6510683788, "environment_steps": 64000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.4221285723669442, "eta_seconds": 1519.662860520999, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_016000.safetensors", "metrics": {"approx_kl": 8.177443310553034e-05, "chaser_grad_norm": 0.026209967210888863, "chaser_new_fields": 82.67500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.415000915527344, "chaser_subspace_norm": 80.78495025634766, "collisions_per_episode": 1.7000000476837158, "entropy": 0.0644609984010458, "explorer_grad_norm": 1.3456061631441116, "explorer_new_fields": 80.45000457763672, "explorer_partner_vision": 0.0, "explorer_return": 69.9625015258789, "explorer_subspace_norm": 26.441322326660156, "final_distance": 5.649790287017822, "max_param_abs": 1.486730933189392, "policy_loss": -0.0005736820357267192, "value_loss": 3.2932186126708984}, "progress_fraction": 0.8, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:17:36.374514+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983856.3745298, "update": 16000}
+{"device": "cuda", "elapsed_hours": 1.7405432201391604, "elapsed_seconds": 6265.955592500977, "env_steps_per_second": 10533.110078052267, "environment_steps": 66000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.3692061376052764, "eta_seconds": 1329.142095378995, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_016000.safetensors", "metrics": {"approx_kl": 1.3870245280732385e-05, "chaser_grad_norm": 0.03339986642822623, "chaser_new_fields": 83.625, "chaser_partner_vision": 0.0, "chaser_return": 6.625000953674316, "chaser_subspace_norm": 81.69694519042969, "collisions_per_episode": 1.5750000476837158, "entropy": 0.05169977340847254, "explorer_grad_norm": 1.3918738961219788, "explorer_new_fields": 81.4749984741211, "explorer_partner_vision": 0.0, "explorer_return": 71.5, "explorer_subspace_norm": 25.642988204956055, "final_distance": 5.413881301879883, "max_param_abs": 1.4971561431884766, "policy_loss": -0.0004798526813694082, "value_loss": 3.2801648378372192}, "progress_fraction": 0.825, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:20:43.678663+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984043.6786747, "update": 16500}
+{"device": "cuda", "elapsed_hours": 1.7924729026744415, "elapsed_seconds": 6452.902449627989, "env_steps_per_second": 10537.893689051538, "environment_steps": 68000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.3163187475307838, "eta_seconds": 1138.7474911108218, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_017000.safetensors", "metrics": {"approx_kl": 0.00021581246380400487, "chaser_grad_norm": 0.023911027936264873, "chaser_new_fields": 82.2750015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.355000972747803, "chaser_subspace_norm": 77.47178649902344, "collisions_per_episode": 1.475000023841858, "entropy": 0.05338282324373722, "explorer_grad_norm": 0.8776888698339462, "explorer_new_fields": 80.5, "explorer_partner_vision": 0.0, "explorer_return": 70.0374984741211, "explorer_subspace_norm": 25.362728118896484, "final_distance": 5.1118950843811035, "max_param_abs": 1.5214029550552368, "policy_loss": -0.00040889874164840023, "value_loss": 3.3093971610069275}, "progress_fraction": 0.85, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:23:50.625521+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984230.6255345, "update": 17000}
+{"device": "cuda", "elapsed_hours": 1.8450617680116557, "elapsed_seconds": 6642.22236484196, "env_steps_per_second": 10538.641459900226, "environment_steps": 70000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.26358025257309364, "eta_seconds": 948.8889092631371, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_017000.safetensors", "metrics": {"approx_kl": 0.0002657267602084179, "chaser_grad_norm": 0.05063078086823225, "chaser_new_fields": 82.7750015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.455001354217529, "chaser_subspace_norm": 75.80194854736328, "collisions_per_episode": 1.350000023841858, "entropy": 0.05619639251381159, "explorer_grad_norm": 0.7586047500371933, "explorer_new_fields": 81.5250015258789, "explorer_partner_vision": 0.0, "explorer_return": 71.80000305175781, "explorer_subspace_norm": 25.632326126098633, "final_distance": 5.240739345550537, "max_param_abs": 1.5486178398132324, "policy_loss": -0.0004507996744940179, "value_loss": 3.1661828756332397}, "progress_fraction": 0.875, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:26:59.945437+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984419.9454508, "update": 17500}
+{"device": "cuda", "elapsed_hours": 1.8987893208727797, "elapsed_seconds": 6835.641555142007, "env_steps_per_second": 10533.02742971347, "environment_steps": 72000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.21097659120808665, "eta_seconds": 759.5157283491119, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_018000.safetensors", "metrics": {"approx_kl": 0.000476315018807405, "chaser_grad_norm": 0.0847174865193665, "chaser_new_fields": 82.42500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.40500020980835, "chaser_subspace_norm": 73.47077178955078, "collisions_per_episode": 1.1750000715255737, "entropy": 0.05018320120871067, "explorer_grad_norm": 1.9407499842345715, "explorer_new_fields": 79.67500305175781, "explorer_partner_vision": 0.0, "explorer_return": 68.8375015258789, "explorer_subspace_norm": 24.79322052001953, "final_distance": 5.659545421600342, "max_param_abs": 1.5856611728668213, "policy_loss": -0.0004987320494791447, "value_loss": 3.22615909576416}, "progress_fraction": 0.9, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:30:13.364629+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984613.3646438, "update": 18000}
+{"device": "cuda", "elapsed_hours": 1.9517543001891722, "elapsed_seconds": 7026.31548068102, "env_steps_per_second": 10531.83566884583, "environment_steps": 74000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.15825034866398693, "eta_seconds": 569.701255190353, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_018000.safetensors", "metrics": {"approx_kl": 0.00022815836740864605, "chaser_grad_norm": 0.0501972446218133, "chaser_new_fields": 81.5250015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.215001106262207, "chaser_subspace_norm": 81.19401550292969, "collisions_per_episode": 1.524999976158142, "entropy": 0.059275402687489986, "explorer_grad_norm": 0.6187672689557076, "explorer_new_fields": 79.95000457763672, "explorer_partner_vision": 0.0, "explorer_return": 69.55000305175781, "explorer_subspace_norm": 26.134788513183594, "final_distance": 6.221502780914307, "max_param_abs": 1.6402511596679688, "policy_loss": -0.0005101506211815376, "value_loss": 3.2488821148872375}, "progress_fraction": 0.925, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:33:24.038552+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984804.0385628, "update": 18500}
+{"device": "cuda", "elapsed_hours": 2.00550099451415, "elapsed_seconds": 7219.80358025094, "env_steps_per_second": 10526.602165173925, "environment_steps": 76000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.10555268392179737, "eta_seconds": 379.9896621184705, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_019000.safetensors", "metrics": {"approx_kl": -1.0564466708995468e-05, "chaser_grad_norm": 0.13344785012304783, "chaser_new_fields": 81.57500457763672, "chaser_partner_vision": 0.0, "chaser_return": 6.220000743865967, "chaser_subspace_norm": 77.87812042236328, "collisions_per_episode": 1.3000000715255737, "entropy": 0.057777862064540386, "explorer_grad_norm": 0.9379898607730865, "explorer_new_fields": 77.6500015258789, "explorer_partner_vision": 0.0, "explorer_return": 65.98750305175781, "explorer_subspace_norm": 25.909114837646484, "final_distance": 5.9693779945373535, "max_param_abs": 1.6938241720199585, "policy_loss": -0.0004634228043762789, "value_loss": 3.289840877056122}, "progress_fraction": 0.95, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:36:37.526654+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984997.5266697, "update": 19000}
+{"device": "cuda", "elapsed_hours": 2.058184965768046, "elapsed_seconds": 7409.465876764967, "env_steps_per_second": 10527.074595835164, "environment_steps": 78000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.05277397348123196, "eta_seconds": 189.98630453243504, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_019000.safetensors", "metrics": {"approx_kl": 8.450402172603677e-05, "chaser_grad_norm": 0.0522395009174943, "chaser_new_fields": 85.05000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.895001411437988, "chaser_subspace_norm": 85.82374572753906, "collisions_per_episode": 1.524999976158142, "entropy": 0.0807732418179512, "explorer_grad_norm": 0.4299086779356003, "explorer_new_fields": 76.42500305175781, "explorer_partner_vision": 0.0, "explorer_return": 64.11250305175781, "explorer_subspace_norm": 25.97504425048828, "final_distance": 6.12801456451416, "max_param_abs": 1.7207889556884766, "policy_loss": -0.001022417795612185, "value_loss": 3.4723560214042664}, "progress_fraction": 0.975, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:39:47.188947+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782985187.1889591, "update": 19500}
+{"device": "cuda", "elapsed_hours": 2.1116820847863953, "elapsed_seconds": 7602.055505231023, "env_steps_per_second": 10523.469599104017, "environment_steps": 80000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.0, "eta_seconds": 0.0, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/update_019000.safetensors", "metrics": {"approx_kl": 0.0002434901476933704, "chaser_grad_norm": 0.07690595369786024, "chaser_new_fields": 83.0999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.540000915527344, "chaser_subspace_norm": 75.278076171875, "collisions_per_episode": 1.25, "entropy": 0.03620169032365084, "explorer_grad_norm": 0.2844117097556591, "explorer_new_fields": 75.5250015258789, "explorer_partner_vision": 0.0, "explorer_return": 62.76250076293945, "explorer_subspace_norm": 28.751102447509766, "final_distance": 5.306502342224121, "max_param_abs": 1.7384926080703735, "policy_loss": -0.0007488939012390006, "value_loss": 3.4187601804733276}, "progress_fraction": 1.0, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:42:59.778578+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782985379.7785916, "update": 20000}
+{"device": "cuda", "elapsed_hours": 2.111703482349177, "elapsed_seconds": 7602.132536457037, "env_steps_per_second": 10523.362966424142, "environment_steps": 80000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.0, "eta_seconds": 0.0, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors", "metrics": {"approx_kl": 0.0002434901476933704, "chaser_grad_norm": 0.07690595369786024, "chaser_new_fields": 83.0999984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.540000915527344, "chaser_subspace_norm": 75.278076171875, "collisions_per_episode": 1.25, "entropy": 0.03620169032365084, "explorer_grad_norm": 0.2844117097556591, "explorer_new_fields": 75.5250015258789, "explorer_partner_vision": 0.0, "explorer_return": 62.76250076293945, "explorer_subspace_norm": 28.751102447509766, "final_distance": 5.306502342224121, "max_param_abs": 1.7384926080703735, "policy_loss": -0.0007488939012390006, "value_loss": 3.4187601804733276}, "progress_fraction": 1.0, "schema_version": 2, "state": "completed", "timestamp": "2026-07-02T09:42:59.855609+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782985379.8556223, "update": 20000}

non_social/seed_0000/attempt_01/status.json

@@ -0,0 +1,40 @@
+{
+  "device": "cuda",
+  "elapsed_hours": 2.111703482349177,
+  "elapsed_seconds": 7602.132536457037,
+  "env_steps_per_second": 10523.362966424142,
+  "environment_steps": 80000000,
+  "error": null,
+  "estimated_cost_usd": null,
+  "eta_hours": 0.0,
+  "eta_seconds": 0.0,
+  "healthy": true,
+  "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0000/attempt_01/checkpoints/latest.safetensors",
+  "metrics": {
+    "approx_kl": 0.0002434901476933704,
+    "chaser_grad_norm": 0.07690595369786024,
+    "chaser_new_fields": 83.0999984741211,
+    "chaser_partner_vision": 0.0,
+    "chaser_return": 6.540000915527344,
+    "chaser_subspace_norm": 75.278076171875,
+    "collisions_per_episode": 1.25,
+    "entropy": 0.03620169032365084,
+    "explorer_grad_norm": 0.2844117097556591,
+    "explorer_new_fields": 75.5250015258789,
+    "explorer_partner_vision": 0.0,
+    "explorer_return": 62.76250076293945,
+    "explorer_subspace_norm": 28.751102447509766,
+    "final_distance": 5.306502342224121,
+    "max_param_abs": 1.7384926080703735,
+    "policy_loss": -0.0007488939012390006,
+    "value_loss": 3.4187601804733276
+  },
+  "progress_fraction": 1.0,
+  "schema_version": 2,
+  "state": "completed",
+  "timestamp": "2026-07-02T09:42:59.855609+00:00",
+  "total_environment_steps": 80000000,
+  "total_updates": 20000,
+  "unix_time": 1782985379.8556223,
+  "update": 20000
+}

non_social/seed_0001/attempt_01/checkpoints/latest.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6c8227e6a0fa824f91e1ee406af15f03ca2f65ee00d67445626986d7698c1c3
+size 11460368

non_social/seed_0001/attempt_01/config.json

@@ -0,0 +1,44 @@
+{
+  "architecture": "mlp",
+  "attempt_id": "attempt_01",
+  "batch_size": 40,
+  "checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors",
+  "checkpoint_every": 1000,
+  "checkpoint_every_seconds": 1800.0,
+  "clip_epsilon": 0.2,
+  "cost_per_hour": null,
+  "cuda_tf32": true,
+  "device": "cuda",
+  "entropy_coef": 0.01,
+  "env": {
+    "grid_size": 10,
+    "max_steps": 100,
+    "partner_visibility": "none",
+    "task": "non_social",
+    "vision_radius": 3
+  },
+  "experiment_id": "mouse-run-run-2-mlp",
+  "finite_guard": true,
+  "fused_agent_rollout": false,
+  "gae_lambda": 0.95,
+  "gamma": 0.99,
+  "grad_clip": 1.0,
+  "hidden_size": 256,
+  "learning_rate": 0.0003,
+  "log_every": 500,
+  "metrics_path": null,
+  "ppo_epochs": 4,
+  "recurrent_l2_coef": 0.3,
+  "resume_from": null,
+  "run_dir": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01",
+  "run_id": "mouse-run-run-2-mlp.non_social.seed_0001.attempt_01",
+  "seed": 1,
+  "status_every_seconds": 1800.0,
+  "status_path": null,
+  "subspace_metric_period": 10,
+  "tensorboard_dir": null,
+  "triton_env_step": true,
+  "updates": 20000,
+  "value_clip": 10.0,
+  "value_coef": 0.5
+}

non_social/seed_0001/attempt_01/metrics.jsonl

@@ -0,0 +1,42 @@
+{"device": "cuda", "elapsed_hours": 0.0006646091572474689, "elapsed_seconds": 2.392592966090888, "env_steps_per_second": 1671.8263644047056, "environment_steps": 4000, "error": null, "estimated_cost_usd": null, "eta_hours": 13.291518535792129, "eta_seconds": 47849.466728851665, "healthy": true, "latest_checkpoint": null, "metrics": {"approx_kl": -5.105864413525174e-06, "chaser_grad_norm": 0.35422321408987045, "chaser_new_fields": 34.47500228881836, "chaser_partner_vision": 0.0, "chaser_return": -3.140000343322754, "chaser_subspace_norm": 0.0020734171848744154, "collisions_per_episode": 1.100000023841858, "entropy": 2.7725194096565247, "explorer_grad_norm": 0.10158612206578255, "explorer_new_fields": 36.20000076293945, "explorer_partner_vision": 0.0, "explorer_return": 4.037499904632568, "explorer_subspace_norm": 0.002020237036049366, "final_distance": 5.169232368469238, "max_param_abs": 0.1542729288339615, "policy_loss": -0.0005274267090716878, "value_loss": 2.9001960158348083}, "progress_fraction": 5e-05, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:36:20.165036+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782977780.1650496, "update": 1}
+{"device": "cuda", "elapsed_hours": 0.05299670888110995, "elapsed_seconds": 190.78815197199583, "env_steps_per_second": 10482.83124149954, "environment_steps": 2000000, "error": null, "estimated_cost_usd": null, "eta_hours": 2.066871646363288, "eta_seconds": 7440.737926907837, "healthy": true, "latest_checkpoint": null, "metrics": {"approx_kl": 0.00024995170555008883, "chaser_grad_norm": 0.2540866546332836, "chaser_new_fields": 76.875, "chaser_partner_vision": 0.0, "chaser_return": 5.330000877380371, "chaser_subspace_norm": 17.4329891204834, "collisions_per_episode": 1.0500000715255737, "entropy": 0.39347945898771286, "explorer_grad_norm": 0.6242107301950455, "explorer_new_fields": 66.0999984741211, "explorer_partner_vision": 0.0, "explorer_return": 48.625, "explorer_subspace_norm": 9.651309967041016, "final_distance": 5.658512592315674, "max_param_abs": 0.3182811737060547, "policy_loss": -0.0018509065173546446, "value_loss": 4.656160950660706}, "progress_fraction": 0.025, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:39:28.560596+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782977968.5606086, "update": 500}
+{"device": "cuda", "elapsed_hours": 0.10618631221250527, "elapsed_seconds": 382.270723965019, "env_steps_per_second": 10463.788486104508, "environment_steps": 4000000, "error": null, "estimated_cost_usd": null, "eta_hours": 2.0175399320376, "eta_seconds": 7263.143755335361, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_001000.safetensors", "metrics": {"approx_kl": 0.00034994851500846913, "chaser_grad_norm": 0.19566128961741924, "chaser_new_fields": 79.25, "chaser_partner_vision": 0.0, "chaser_return": 5.7950005531311035, "chaser_subspace_norm": 29.966073989868164, "collisions_per_episode": 1.0750000476837158, "entropy": 0.26417846977710724, "explorer_grad_norm": 0.7438876032829285, "explorer_new_fields": 66.95000457763672, "explorer_partner_vision": 0.0, "explorer_return": 50.275001525878906, "explorer_subspace_norm": 10.79336166381836, "final_distance": 5.03670072555542, "max_param_abs": 0.3332476317882538, "policy_loss": -0.0016920956687584976, "value_loss": 4.247400283813477}, "progress_fraction": 0.05, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:42:40.043166+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978160.0431795, "update": 1000}
+{"device": "cuda", "elapsed_hours": 0.16047670955973445, "elapsed_seconds": 577.716154415044, "env_steps_per_second": 10385.72308242824, "environment_steps": 6000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.9792127512367248, "eta_seconds": 7125.165904452209, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_001000.safetensors", "metrics": {"approx_kl": 0.0008375835176397395, "chaser_grad_norm": 0.18825266137719154, "chaser_new_fields": 81.625, "chaser_partner_vision": 0.0, "chaser_return": 6.27000093460083, "chaser_subspace_norm": 40.897972106933594, "collisions_per_episode": 1.25, "entropy": 0.256906658411026, "explorer_grad_norm": 0.5816270112991333, "explorer_new_fields": 65.4749984741211, "explorer_partner_vision": 0.0, "explorer_return": 47.72500228881836, "explorer_subspace_norm": 11.335468292236328, "final_distance": 4.564890384674072, "max_param_abs": 0.4130561649799347, "policy_loss": -0.0011627619122265287, "value_loss": 4.187922835350037}, "progress_fraction": 0.075, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:45:55.488596+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978355.488607, "update": 1500}
+{"device": "cuda", "elapsed_hours": 0.21357180113195984, "elapsed_seconds": 768.8584840750555, "env_steps_per_second": 10405.035732452223, "environment_steps": 8000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.9221462101876385, "eta_seconds": 6919.726356675499, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_002000.safetensors", "metrics": {"approx_kl": 0.0005999028207475943, "chaser_grad_norm": 0.20418540574610233, "chaser_new_fields": 81.55000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.260001182556152, "chaser_subspace_norm": 47.05528259277344, "collisions_per_episode": 1.3000000715255737, "entropy": 0.20487841591238976, "explorer_grad_norm": 0.7134600207209587, "explorer_new_fields": 67.7249984741211, "explorer_partner_vision": 0.0, "explorer_return": 50.95000076293945, "explorer_subspace_norm": 11.689831733703613, "final_distance": 5.1953253746032715, "max_param_abs": 0.46658068895339966, "policy_loss": -0.0004994259075061791, "value_loss": 3.8595205545425415}, "progress_fraction": 0.1, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:49:06.630927+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978546.6309407, "update": 2000}
+{"device": "cuda", "elapsed_hours": 0.26611383178057924, "elapsed_seconds": 958.0097944100853, "env_steps_per_second": 10438.306641904126, "environment_steps": 10000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.8627968224640548, "eta_seconds": 6706.068560870597, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_002000.safetensors", "metrics": {"approx_kl": 0.0002379925559409246, "chaser_grad_norm": 0.15856351889669895, "chaser_new_fields": 81.1500015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.160001277923584, "chaser_subspace_norm": 47.54892349243164, "collisions_per_episode": 1.3000000715255737, "entropy": 0.19381681457161903, "explorer_grad_norm": 1.4128717184066772, "explorer_new_fields": 68.55000305175781, "explorer_partner_vision": 0.0, "explorer_return": 52.45000076293945, "explorer_subspace_norm": 14.399904251098633, "final_distance": 4.250465393066406, "max_param_abs": 0.4694162905216217, "policy_loss": -0.0007240931175118703, "value_loss": 4.089799165725708}, "progress_fraction": 0.125, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:52:15.782237+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978735.7822506, "update": 2500}
+{"device": "cuda", "elapsed_hours": 0.31946793677974106, "elapsed_seconds": 1150.0845724070678, "env_steps_per_second": 10434.01527844567, "environment_steps": 12000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.8103183084185326, "eta_seconds": 6517.145910306717, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_003000.safetensors", "metrics": {"approx_kl": 0.0005075695641181543, "chaser_grad_norm": 0.0860081585124135, "chaser_new_fields": 82.875, "chaser_partner_vision": 0.0, "chaser_return": 6.490001201629639, "chaser_subspace_norm": 47.72903060913086, "collisions_per_episode": 1.8000000715255737, "entropy": 0.1328066848218441, "explorer_grad_norm": 0.5817417502403259, "explorer_new_fields": 71.17500305175781, "explorer_partner_vision": 0.0, "explorer_return": 56.38750076293945, "explorer_subspace_norm": 14.815073013305664, "final_distance": 4.913993835449219, "max_param_abs": 0.5209510326385498, "policy_loss": -0.00031853549470550746, "value_loss": 3.6523645520210266}, "progress_fraction": 0.15, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:55:27.857015+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782978927.8570433, "update": 3000}
+{"device": "cuda", "elapsed_hours": 0.37271470920835986, "elapsed_seconds": 1341.7729531500954, "env_steps_per_second": 10433.956033419845, "environment_steps": 14000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.7570836291251248, "eta_seconds": 6325.501064850449, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_003000.safetensors", "metrics": {"approx_kl": 0.00013817141942573663, "chaser_grad_norm": 0.23563339933753014, "chaser_new_fields": 83.05000305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.530001163482666, "chaser_subspace_norm": 46.80853271484375, "collisions_per_episode": 1.375, "entropy": 0.10780291259288788, "explorer_grad_norm": 0.5218307226896286, "explorer_new_fields": 71.0250015258789, "explorer_partner_vision": 0.0, "explorer_return": 56.04999923706055, "explorer_subspace_norm": 16.33576202392578, "final_distance": 4.8094482421875, "max_param_abs": 0.637711226940155, "policy_loss": -0.0008312481108596614, "value_loss": 3.6552406549453735}, "progress_fraction": 0.175, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T07:58:39.545395+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979119.5454068, "update": 3500}
+{"device": "cuda", "elapsed_hours": 0.42564808120805536, "elapsed_seconds": 1532.3330923489993, "env_steps_per_second": 10441.593984942727, "environment_steps": 16000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.7025923248322214, "eta_seconds": 6129.332369395997, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_004000.safetensors", "metrics": {"approx_kl": -5.299170846001294e-05, "chaser_grad_norm": 0.1636909916996956, "chaser_new_fields": 81.92500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.30000114440918, "chaser_subspace_norm": 55.12071228027344, "collisions_per_episode": 1.725000023841858, "entropy": 0.12000446207821369, "explorer_grad_norm": 1.3998910635709763, "explorer_new_fields": 71.9000015258789, "explorer_partner_vision": 0.0, "explorer_return": 57.28750228881836, "explorer_subspace_norm": 16.26192283630371, "final_distance": 4.884814739227295, "max_param_abs": 0.7220480442047119, "policy_loss": -0.0007309603997214253, "value_loss": 3.6593281626701355}, "progress_fraction": 0.2, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:01:50.105535+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979310.105551, "update": 4000}
+{"device": "cuda", "elapsed_hours": 0.47814068301862184, "elapsed_seconds": 1721.3064588670386, "env_steps_per_second": 10457.173333241062, "environment_steps": 18000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.6469290192863644, "eta_seconds": 5928.944469430911, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_004000.safetensors", "metrics": {"approx_kl": 7.421613951296563e-05, "chaser_grad_norm": 0.23856427893042564, "chaser_new_fields": 81.125, "chaser_partner_vision": 0.0, "chaser_return": 6.140000820159912, "chaser_subspace_norm": 55.25377655029297, "collisions_per_episode": 1.475000023841858, "entropy": 0.1395217962563038, "explorer_grad_norm": 1.1217033490538597, "explorer_new_fields": 68.57500457763672, "explorer_partner_vision": 0.0, "explorer_return": 52.26250076293945, "explorer_subspace_norm": 17.047136306762695, "final_distance": 4.939172744750977, "max_param_abs": 0.8441739678382874, "policy_loss": -0.00040146764449322525, "value_loss": 3.8861085176467896}, "progress_fraction": 0.225, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:04:59.078901+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979499.078913, "update": 4500}
+{"device": "cuda", "elapsed_hours": 0.5305872613688957, "elapsed_seconds": 1910.1141409280244, "env_steps_per_second": 10470.57847039604, "environment_steps": 20000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.5917617841066871, "eta_seconds": 5730.342422784073, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_005000.safetensors", "metrics": {"approx_kl": 0.0001957275774418577, "chaser_grad_norm": 0.06718951277434826, "chaser_new_fields": 79.67500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.835000514984131, "chaser_subspace_norm": 61.439979553222656, "collisions_per_episode": 1.1750000715255737, "entropy": 0.09594621136784554, "explorer_grad_norm": 0.5464240312576294, "explorer_new_fields": 68.9749984741211, "explorer_partner_vision": 0.0, "explorer_return": 53.01250076293945, "explorer_subspace_norm": 17.778528213500977, "final_distance": 5.9304118156433105, "max_param_abs": 0.9190318584442139, "policy_loss": -0.0004385370368851227, "value_loss": 3.810237228870392}, "progress_fraction": 0.25, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:08:07.886584+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979687.8865972, "update": 5000}
+{"device": "cuda", "elapsed_hours": 0.5838003536613865, "elapsed_seconds": 2101.6812731809914, "env_steps_per_second": 10467.809881896119, "environment_steps": 22000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.53911002328911, "eta_seconds": 5540.796083840796, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_005000.safetensors", "metrics": {"approx_kl": 0.0007284536260159166, "chaser_grad_norm": 0.19790029153227806, "chaser_new_fields": 79.92500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.900001049041748, "chaser_subspace_norm": 60.71714782714844, "collisions_per_episode": 1.6750000715255737, "entropy": 0.09285321645438671, "explorer_grad_norm": 1.190895400941372, "explorer_new_fields": 71.57500457763672, "explorer_partner_vision": 0.0, "explorer_return": 56.95000076293945, "explorer_subspace_norm": 19.046768188476562, "final_distance": 4.46820592880249, "max_param_abs": 0.9952329993247986, "policy_loss": -0.0007394702993295255, "value_loss": 3.7340222597122192}, "progress_fraction": 0.275, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:11:19.453715+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782979879.4537265, "update": 5500}
+{"device": "cuda", "elapsed_hours": 0.637212698103071, "elapsed_seconds": 2293.9657131710555, "env_steps_per_second": 10462.231349928803, "environment_steps": 24000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.4868296289071656, "eta_seconds": 5352.586664065796, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_006000.safetensors", "metrics": {"approx_kl": 0.00012293299168864014, "chaser_grad_norm": 0.1078545544296503, "chaser_new_fields": 79.125, "chaser_partner_vision": 0.0, "chaser_return": 5.755000591278076, "chaser_subspace_norm": 68.20173645019531, "collisions_per_episode": 1.25, "entropy": 0.1013022381812334, "explorer_grad_norm": 0.5451182201504707, "explorer_new_fields": 70.57500457763672, "explorer_partner_vision": 0.0, "explorer_return": 55.41250228881836, "explorer_subspace_norm": 17.95201301574707, "final_distance": 5.370412349700928, "max_param_abs": 1.0787969827651978, "policy_loss": -0.0004317292224147451, "value_loss": 3.706162452697754}, "progress_fraction": 0.3, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:14:31.738157+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980071.738171, "update": 6000}
+{"device": "cuda", "elapsed_hours": 0.6894948651758346, "elapsed_seconds": 2482.1815146330046, "env_steps_per_second": 10474.657009056065, "environment_steps": 26000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.4320277969036563, "eta_seconds": 5155.300068853163, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_006000.safetensors", "metrics": {"approx_kl": 0.0002024353854959693, "chaser_grad_norm": 0.2383432388305664, "chaser_new_fields": 76.92500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.325000762939453, "chaser_subspace_norm": 65.47434997558594, "collisions_per_episode": 1.399999976158142, "entropy": 0.09009139984846115, "explorer_grad_norm": 2.05173921585083, "explorer_new_fields": 74.0, "explorer_partner_vision": 0.0, "explorer_return": 60.400001525878906, "explorer_subspace_norm": 18.082992553710938, "final_distance": 5.4761481285095215, "max_param_abs": 1.11553156375885, "policy_loss": -0.0003861995520466621, "value_loss": 3.5980712175369263}, "progress_fraction": 0.325, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:17:39.953957+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980259.9539683, "update": 6500}
+{"device": "cuda", "elapsed_hours": 0.741385722357243, "elapsed_seconds": 2668.9886004860746, "env_steps_per_second": 10490.865339365128, "environment_steps": 28000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.3768591986634513, "eta_seconds": 4956.693115188425, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_007000.safetensors", "metrics": {"approx_kl": 0.00014662955988614668, "chaser_grad_norm": 0.13823741115629673, "chaser_new_fields": 76.375, "chaser_partner_vision": 0.0, "chaser_return": 5.185000896453857, "chaser_subspace_norm": 81.42784118652344, "collisions_per_episode": 1.1750000715255737, "entropy": 0.08847093023359776, "explorer_grad_norm": 0.4604616239666939, "explorer_new_fields": 75.42500305175781, "explorer_partner_vision": 0.0, "explorer_return": 62.61249923706055, "explorer_subspace_norm": 21.18718147277832, "final_distance": 6.204265117645264, "max_param_abs": 1.158452033996582, "policy_loss": -0.0010841385633639788, "value_loss": 4.00962769985199}, "progress_fraction": 0.35, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:20:46.761043+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980446.7610588, "update": 7000}
+{"device": "cuda", "elapsed_hours": 0.7941792751552809, "elapsed_seconds": 2859.045390559011, "env_steps_per_second": 10493.012842350952, "environment_steps": 30000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.3236321252588015, "eta_seconds": 4765.075650931685, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_007000.safetensors", "metrics": {"approx_kl": 2.783902959220541e-05, "chaser_grad_norm": 0.07786290813237429, "chaser_new_fields": 83.6500015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.655002117156982, "chaser_subspace_norm": 94.4039306640625, "collisions_per_episode": 1.024999976158142, "entropy": 0.09832780249416828, "explorer_grad_norm": 1.3362184911966324, "explorer_new_fields": 75.625, "explorer_partner_vision": 0.0, "explorer_return": 63.0625, "explorer_subspace_norm": 21.0552921295166, "final_distance": 5.50551176071167, "max_param_abs": 1.0359666347503662, "policy_loss": -0.00043181558649152407, "value_loss": 3.7632511854171753}, "progress_fraction": 0.375, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:23:56.817832+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980636.8178422, "update": 7500}
+{"device": "cuda", "elapsed_hours": 0.8470003187508296, "elapsed_seconds": 3049.2011475029867, "env_steps_per_second": 10494.551999695079, "environment_steps": 32000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.2705004781262446, "eta_seconds": 4573.80172125448, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_008000.safetensors", "metrics": {"approx_kl": 0.0004685354021869981, "chaser_grad_norm": 0.15229122154414654, "chaser_new_fields": 82.42500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.4500017166137695, "chaser_subspace_norm": 96.91693878173828, "collisions_per_episode": 1.0500000715255737, "entropy": 0.06496398337185383, "explorer_grad_norm": 1.082899585366249, "explorer_new_fields": 76.20000457763672, "explorer_partner_vision": 0.0, "explorer_return": 63.70000076293945, "explorer_subspace_norm": 23.27690887451172, "final_distance": 5.564136505126953, "max_param_abs": 0.9712896347045898, "policy_loss": -0.0011512674705596027, "value_loss": 3.5662338733673096}, "progress_fraction": 0.4, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:27:06.973591+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782980826.9736054, "update": 8000}
+{"device": "cuda", "elapsed_hours": 0.8991850248044486, "elapsed_seconds": 3237.066089296015, "env_steps_per_second": 10503.3382273002, "environment_steps": 34000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.2165444453236658, "eta_seconds": 4379.560003165197, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_008000.safetensors", "metrics": {"approx_kl": 0.0002205174978130442, "chaser_grad_norm": 0.08294959180057049, "chaser_new_fields": 81.2249984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.190001010894775, "chaser_subspace_norm": 101.9328384399414, "collisions_per_episode": 1.0, "entropy": 0.045907335355877876, "explorer_grad_norm": 0.5273615941405296, "explorer_new_fields": 77.17500305175781, "explorer_partner_vision": 0.0, "explorer_return": 65.1624984741211, "explorer_subspace_norm": 22.970808029174805, "final_distance": 5.270810127258301, "max_param_abs": 0.974787712097168, "policy_loss": -0.0005561281775072757, "value_loss": 3.466669023036957}, "progress_fraction": 0.425, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:30:14.838531+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981014.8385422, "update": 8500}
+{"device": "cuda", "elapsed_hours": 0.9520237867916714, "elapsed_seconds": 3427.285632450017, "env_steps_per_second": 10503.93922792632, "environment_steps": 36000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.163584628300932, "eta_seconds": 4188.9046618833545, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_009000.safetensors", "metrics": {"approx_kl": 5.521597716218185e-05, "chaser_grad_norm": 0.13706197775900364, "chaser_new_fields": 83.70000457763672, "chaser_partner_vision": 0.0, "chaser_return": 6.675000190734863, "chaser_subspace_norm": 95.42662811279297, "collisions_per_episode": 0.8500000238418579, "entropy": 0.04296411946415901, "explorer_grad_norm": 0.41673433035612106, "explorer_new_fields": 77.375, "explorer_partner_vision": 0.0, "explorer_return": 65.76250457763672, "explorer_subspace_norm": 25.36161994934082, "final_distance": 5.151485443115234, "max_param_abs": 0.9597460031509399, "policy_loss": -0.0002633670809331079, "value_loss": 3.5941293239593506}, "progress_fraction": 0.45, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:33:25.058075+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981205.058089, "update": 9000}
+{"device": "cuda", "elapsed_hours": 1.0056366006861208, "elapsed_seconds": 3620.291762470035, "env_steps_per_second": 10496.391587531483, "environment_steps": 38000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.1114930849688704, "eta_seconds": 4001.3751058879334, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_009000.safetensors", "metrics": {"approx_kl": 0.00012517708195831923, "chaser_grad_norm": 0.2932741940021515, "chaser_new_fields": 72.20000457763672, "chaser_partner_vision": 0.0, "chaser_return": 4.380000591278076, "chaser_subspace_norm": 89.20765686035156, "collisions_per_episode": 1.0, "entropy": 0.0684291161596775, "explorer_grad_norm": 1.9711225181818008, "explorer_new_fields": 73.2750015258789, "explorer_partner_vision": 0.0, "explorer_return": 59.462501525878906, "explorer_subspace_norm": 24.477724075317383, "final_distance": 5.2902936935424805, "max_param_abs": 0.9854035973548889, "policy_loss": -0.0001851809129824744, "value_loss": 3.8698012232780457}, "progress_fraction": 0.475, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:36:38.064206+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981398.0642197, "update": 9500}
+{"device": "cuda", "elapsed_hours": 1.0596000968719534, "elapsed_seconds": 3814.5603487390326, "env_steps_per_second": 10486.13636777897, "environment_steps": 40000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.0596000968719534, "eta_seconds": 3814.560348739032, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_010000.safetensors", "metrics": {"approx_kl": 0.00038964367026812774, "chaser_grad_norm": 0.09080993384122849, "chaser_new_fields": 80.67500305175781, "chaser_partner_vision": 0.0, "chaser_return": 6.085000991821289, "chaser_subspace_norm": 79.93997192382812, "collisions_per_episode": 1.024999976158142, "entropy": 0.05338955298066139, "explorer_grad_norm": 0.8891770392656326, "explorer_new_fields": 75.32500457763672, "explorer_partner_vision": 0.0, "explorer_return": 62.3125, "explorer_subspace_norm": 26.096323013305664, "final_distance": 5.22806978225708, "max_param_abs": 0.9808329343795776, "policy_loss": -0.0008176689683470073, "value_loss": 3.6155094504356384}, "progress_fraction": 0.5, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:39:52.332791+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981592.3328056, "update": 10000}
+{"device": "cuda", "elapsed_hours": 1.113099402428359, "elapsed_seconds": 4007.157848742092, "env_steps_per_second": 10481.244209829278, "environment_steps": 42000000, "error": null, "estimated_cost_usd": null, "eta_hours": 1.00708993553042, "eta_seconds": 3625.523767909512, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_010000.safetensors", "metrics": {"approx_kl": 0.00045353863636776737, "chaser_grad_norm": 0.06808555126190186, "chaser_new_fields": 78.92500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.690001010894775, "chaser_subspace_norm": 90.07524871826172, "collisions_per_episode": 1.4250000715255737, "entropy": 0.056026156060397625, "explorer_grad_norm": 0.768612265586853, "explorer_new_fields": 73.1500015258789, "explorer_partner_vision": 0.0, "explorer_return": 58.97500228881836, "explorer_subspace_norm": 24.316312789916992, "final_distance": 4.746428966522217, "max_param_abs": 0.9814541339874268, "policy_loss": -0.0005006787593009321, "value_loss": 3.4611494541168213}, "progress_fraction": 0.525, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:43:04.930293+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981784.9303074, "update": 10500}
+{"device": "cuda", "elapsed_hours": 1.166562008252238, "elapsed_seconds": 4199.623229708057, "env_steps_per_second": 10477.130350347816, "environment_steps": 44000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.9544598249336493, "eta_seconds": 3436.0553697611376, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_011000.safetensors", "metrics": {"approx_kl": 0.0002189835984244791, "chaser_grad_norm": 0.14183006435632706, "chaser_new_fields": 80.7249984741211, "chaser_partner_vision": 0.0, "chaser_return": 6.065001010894775, "chaser_subspace_norm": 91.1957015991211, "collisions_per_episode": 2.2250001430511475, "entropy": 0.09379604086279869, "explorer_grad_norm": 0.3437352776527405, "explorer_new_fields": 73.05000305175781, "explorer_partner_vision": 0.0, "explorer_return": 58.97500228881836, "explorer_subspace_norm": 25.498416900634766, "final_distance": 4.4637041091918945, "max_param_abs": 0.9716539978981018, "policy_loss": -0.0009403482038745636, "value_loss": 3.6964353919029236}, "progress_fraction": 0.55, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:46:17.395673+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782981977.3956897, "update": 11000}
+{"device": "cuda", "elapsed_hours": 1.2201823734194557, "elapsed_seconds": 4392.656544310041, "env_steps_per_second": 10472.02291733584, "environment_steps": 46000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.9018739281795977, "eta_seconds": 3246.7461414465515, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_011000.safetensors", "metrics": {"approx_kl": 0.0007552657056391077, "chaser_grad_norm": 0.058179725892841816, "chaser_new_fields": 78.4749984741211, "chaser_partner_vision": 0.0, "chaser_return": 5.590001106262207, "chaser_subspace_norm": 84.99630737304688, "collisions_per_episode": 1.399999976158142, "entropy": 0.07308833859860897, "explorer_grad_norm": 1.2959906607866287, "explorer_new_fields": 77.5, "explorer_partner_vision": 0.0, "explorer_return": 65.57500457763672, "explorer_subspace_norm": 25.410776138305664, "final_distance": 5.0258612632751465, "max_param_abs": 1.0161798000335693, "policy_loss": -0.0010617657257228075, "value_loss": 3.5079697370529175}, "progress_fraction": 0.575, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:49:30.428988+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982170.4290025, "update": 11500}
+{"device": "cuda", "elapsed_hours": 1.2742603790403033, "elapsed_seconds": 4587.337364545092, "env_steps_per_second": 10463.586212556653, "environment_steps": 48000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.8495069193602024, "eta_seconds": 3058.2249096967284, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_012000.safetensors", "metrics": {"approx_kl": 0.00021119906125931925, "chaser_grad_norm": 0.13109773769974709, "chaser_new_fields": 76.9749984741211, "chaser_partner_vision": 0.0, "chaser_return": 5.325000762939453, "chaser_subspace_norm": 82.43509674072266, "collisions_per_episode": 1.600000023841858, "entropy": 0.07194264605641365, "explorer_grad_norm": 1.371770866215229, "explorer_new_fields": 70.0999984741211, "explorer_partner_vision": 0.0, "explorer_return": 54.36249923706055, "explorer_subspace_norm": 23.883390426635742, "final_distance": 5.147349834442139, "max_param_abs": 1.032204270362854, "policy_loss": -0.0007983786445109331, "value_loss": 3.6320714354515076}, "progress_fraction": 0.6, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:52:45.109808+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982365.109821, "update": 12000}
+{"device": "cuda", "elapsed_hours": 1.326985051559459, "elapsed_seconds": 4777.146185614052, "env_steps_per_second": 10466.49988450647, "environment_steps": 50000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.7961910309356753, "eta_seconds": 2866.2877113684312, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_012000.safetensors", "metrics": {"approx_kl": 9.733648125020977e-05, "chaser_grad_norm": 0.21115468814969063, "chaser_new_fields": 76.125, "chaser_partner_vision": 0.0, "chaser_return": 5.125000953674316, "chaser_subspace_norm": 90.5826187133789, "collisions_per_episode": 1.899999976158142, "entropy": 0.07294890843331814, "explorer_grad_norm": 0.7679236233234406, "explorer_new_fields": 70.92500305175781, "explorer_partner_vision": 0.0, "explorer_return": 55.60000228881836, "explorer_subspace_norm": 23.094196319580078, "final_distance": 4.849133491516113, "max_param_abs": 1.0736829042434692, "policy_loss": -0.0006347609012173905, "value_loss": 3.693317234516144}, "progress_fraction": 0.625, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:55:54.918628+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982554.9186392, "update": 12500}
+{"device": "cuda", "elapsed_hours": 1.3789171851091668, "elapsed_seconds": 4964.101866393001, "env_steps_per_second": 10475.208083871184, "environment_steps": 52000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.7424938689049361, "eta_seconds": 2672.97792805777, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_013000.safetensors", "metrics": {"approx_kl": 0.0007749467855087122, "chaser_grad_norm": 0.05683001130819321, "chaser_new_fields": 77.55000305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.455001354217529, "chaser_subspace_norm": 94.65978240966797, "collisions_per_episode": 1.149999976158142, "entropy": 0.0547284884378314, "explorer_grad_norm": 0.8309146612882614, "explorer_new_fields": 74.5999984741211, "explorer_partner_vision": 0.0, "explorer_return": 61.1875, "explorer_subspace_norm": 24.733530044555664, "final_distance": 5.075489044189453, "max_param_abs": 1.0736836194992065, "policy_loss": -0.0010926405941162898, "value_loss": 3.685732841491699}, "progress_fraction": 0.65, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T08:59:01.874310+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982741.8743246, "update": 13000}
+{"device": "cuda", "elapsed_hours": 1.432759276984725, "elapsed_seconds": 5157.9333971450105, "env_steps_per_second": 10469.309283809242, "environment_steps": 54000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.6898470592889417, "eta_seconds": 2483.4494134401903, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_013000.safetensors", "metrics": {"approx_kl": 0.0005101998941146879, "chaser_grad_norm": 0.09003976918756962, "chaser_new_fields": 77.5250015258789, "chaser_partner_vision": 0.0, "chaser_return": 5.4250006675720215, "chaser_subspace_norm": 89.87812042236328, "collisions_per_episode": 1.7000000476837158, "entropy": 0.11065679043531418, "explorer_grad_norm": 0.5014339312911034, "explorer_new_fields": 71.25, "explorer_partner_vision": 0.0, "explorer_return": 56.275001525878906, "explorer_subspace_norm": 21.806711196899414, "final_distance": 5.196125030517578, "max_param_abs": 1.0736504793167114, "policy_loss": -0.0009785575692120574, "value_loss": 3.7667288780212402}, "progress_fraction": 0.675, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:02:15.705839+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782982935.7058506, "update": 13500}
+{"device": "cuda", "elapsed_hours": 1.4857316654825181, "elapsed_seconds": 5348.633995737066, "env_steps_per_second": 10469.962993286279, "environment_steps": 56000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.6367421423496507, "eta_seconds": 2292.2717124587425, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_014000.safetensors", "metrics": {"approx_kl": 0.0002474860123922312, "chaser_grad_norm": 0.05047455057501793, "chaser_new_fields": 77.4000015258789, "chaser_partner_vision": 0.0, "chaser_return": 5.3600006103515625, "chaser_subspace_norm": 87.39171600341797, "collisions_per_episode": 2.0250000953674316, "entropy": 0.08761332742869854, "explorer_grad_norm": 1.0886653810739517, "explorer_new_fields": 71.5, "explorer_partner_vision": 0.0, "explorer_return": 56.57500076293945, "explorer_subspace_norm": 22.313016891479492, "final_distance": 4.741565704345703, "max_param_abs": 1.0863597393035889, "policy_loss": -0.0005897007421591383, "value_loss": 3.664694607257843}, "progress_fraction": 0.7, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:05:26.406439+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983126.4064524, "update": 14000}
+{"device": "cuda", "elapsed_hours": 1.5400618418230765, "elapsed_seconds": 5544.222630563076, "env_steps_per_second": 10461.340365422784, "environment_steps": 58000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.5841613882777188, "eta_seconds": 2102.9809977997875, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_014000.safetensors", "metrics": {"approx_kl": 0.0008129233688594084, "chaser_grad_norm": 0.08111646585166454, "chaser_new_fields": 76.5250015258789, "chaser_partner_vision": 0.0, "chaser_return": 5.185000896453857, "chaser_subspace_norm": 86.90794372558594, "collisions_per_episode": 1.9250000715255737, "entropy": 0.10414018481969833, "explorer_grad_norm": 0.7238684520125389, "explorer_new_fields": 68.42500305175781, "explorer_partner_vision": 0.0, "explorer_return": 51.85000228881836, "explorer_subspace_norm": 22.803674697875977, "final_distance": 5.295478820800781, "max_param_abs": 1.1287198066711426, "policy_loss": -0.0016683751654227308, "value_loss": 3.6987728476524353}, "progress_fraction": 0.725, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:08:41.995072+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983321.9950838, "update": 14500}
+{"device": "cuda", "elapsed_hours": 1.5932511207977789, "elapsed_seconds": 5735.704034872004, "env_steps_per_second": 10460.790800085093, "environment_steps": 60000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.5310837069325929, "eta_seconds": 1911.9013449573345, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_015000.safetensors", "metrics": {"approx_kl": 0.0002415505494585135, "chaser_grad_norm": 0.061741966754198074, "chaser_new_fields": 77.92500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.490001201629639, "chaser_subspace_norm": 90.45314025878906, "collisions_per_episode": 1.774999976158142, "entropy": 0.06925754249095917, "explorer_grad_norm": 0.8019117414951324, "explorer_new_fields": 72.9000015258789, "explorer_partner_vision": 0.0, "explorer_return": 58.60000228881836, "explorer_subspace_norm": 25.05002784729004, "final_distance": 5.04298210144043, "max_param_abs": 1.1126190423965454, "policy_loss": -0.000777919802583682, "value_loss": 3.715272307395935}, "progress_fraction": 0.75, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:11:53.476478+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983513.476493, "update": 15000}
+{"device": "cuda", "elapsed_hours": 1.6452307465797316, "elapsed_seconds": 5922.830687687034, "env_steps_per_second": 10467.967644067849, "environment_steps": 62000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.4776476361037931, "eta_seconds": 1719.5314899736552, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_015000.safetensors", "metrics": {"approx_kl": 0.0014523894221247918, "chaser_grad_norm": 0.07097857166081667, "chaser_new_fields": 79.30000305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.785001277923584, "chaser_subspace_norm": 94.98880767822266, "collisions_per_episode": 1.350000023841858, "entropy": 0.07164299860596657, "explorer_grad_norm": 0.6239117234945297, "explorer_new_fields": 73.92500305175781, "explorer_partner_vision": 0.0, "explorer_return": 60.28750228881836, "explorer_subspace_norm": 25.258012771606445, "final_distance": 5.408947467803955, "max_param_abs": 1.1395074129104614, "policy_loss": -0.0011831774853590105, "value_loss": 3.4303807616233826}, "progress_fraction": 0.775, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:15:00.603130+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983700.6031415, "update": 15500}
+{"device": "cuda", "elapsed_hours": 1.6975080731550245, "elapsed_seconds": 6111.029063358088, "env_steps_per_second": 10472.867881409025, "environment_steps": 64000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.4243770182887561, "eta_seconds": 1527.757265839522, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_016000.safetensors", "metrics": {"approx_kl": 0.0002821423777086274, "chaser_grad_norm": 0.0475468123331666, "chaser_new_fields": 77.2750015258789, "chaser_partner_vision": 0.0, "chaser_return": 5.400001049041748, "chaser_subspace_norm": 101.9891128540039, "collisions_per_episode": 1.350000023841858, "entropy": 0.04990239441394806, "explorer_grad_norm": 0.5156930722296238, "explorer_new_fields": 71.95000457763672, "explorer_partner_vision": 0.0, "explorer_return": 57.51250076293945, "explorer_subspace_norm": 27.15218734741211, "final_distance": 5.17217493057251, "max_param_abs": 1.1514462232589722, "policy_loss": -0.000498238339438295, "value_loss": 3.62851220369339}, "progress_fraction": 0.8, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:18:08.801506+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782983888.8015206, "update": 16000}
+{"device": "cuda", "elapsed_hours": 1.74950597281917, "elapsed_seconds": 6298.221502149012, "env_steps_per_second": 10479.148752942427, "environment_steps": 66000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.37110732756770276, "eta_seconds": 1335.98637924373, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_016000.safetensors", "metrics": {"approx_kl": 0.0001662679765182147, "chaser_grad_norm": 0.058657112531363964, "chaser_new_fields": 77.92500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.4850006103515625, "chaser_subspace_norm": 94.71659088134766, "collisions_per_episode": 1.9250000715255737, "entropy": 0.06443462520837784, "explorer_grad_norm": 0.46494172513484955, "explorer_new_fields": 69.4000015258789, "explorer_partner_vision": 0.0, "explorer_return": 53.275001525878906, "explorer_subspace_norm": 27.593393325805664, "final_distance": 5.075628757476807, "max_param_abs": 1.1629002094268799, "policy_loss": -0.0009729938061013854, "value_loss": 3.6678024530410767}, "progress_fraction": 0.825, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:21:15.993944+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984075.9939582, "update": 16500}
+{"device": "cuda", "elapsed_hours": 1.8016514310583525, "elapsed_seconds": 6485.945151810069, "env_steps_per_second": 10484.208300932496, "environment_steps": 68000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.3179384878338269, "eta_seconds": 1144.5785562017768, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_017000.safetensors", "metrics": {"approx_kl": 0.00023642260933431913, "chaser_grad_norm": 0.12356526777148247, "chaser_new_fields": 77.875, "chaser_partner_vision": 0.0, "chaser_return": 5.445001125335693, "chaser_subspace_norm": 79.68389129638672, "collisions_per_episode": 1.875, "entropy": 0.08755725808441639, "explorer_grad_norm": 0.9381912797689438, "explorer_new_fields": 72.5, "explorer_partner_vision": 0.0, "explorer_return": 58.0, "explorer_subspace_norm": 26.78290557861328, "final_distance": 5.436651706695557, "max_param_abs": 1.267048954963684, "policy_loss": -0.000974916765464684, "value_loss": 3.7329800724983215}, "progress_fraction": 0.85, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:24:23.717595+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984263.71761, "update": 17000}
+{"device": "cuda", "elapsed_hours": 1.854691653937787, "elapsed_seconds": 6676.889954176033, "env_steps_per_second": 10483.922976178272, "environment_steps": 70000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.264955950562541, "eta_seconds": 953.8414220251476, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_017000.safetensors", "metrics": {"approx_kl": 1.806048938457483e-07, "chaser_grad_norm": 0.050238268449902534, "chaser_new_fields": 80.17500305175781, "chaser_partner_vision": 0.0, "chaser_return": 5.945001125335693, "chaser_subspace_norm": 91.65946197509766, "collisions_per_episode": 1.8250000476837158, "entropy": 0.03836974501609802, "explorer_grad_norm": 0.7704875618219376, "explorer_new_fields": 74.20000457763672, "explorer_partner_vision": 0.0, "explorer_return": 60.32500076293945, "explorer_subspace_norm": 26.297279357910156, "final_distance": 5.851404666900635, "max_param_abs": 1.2902907133102417, "policy_loss": -0.00034309800324283657, "value_loss": 3.4310280084609985}, "progress_fraction": 0.875, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:27:34.662396+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984454.6624086, "update": 17500}
+{"device": "cuda", "elapsed_hours": 1.9075580443783353, "elapsed_seconds": 6867.208959762007, "env_steps_per_second": 10484.608874126245, "environment_steps": 72000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.21195089381981505, "eta_seconds": 763.0232177513342, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_018000.safetensors", "metrics": {"approx_kl": 0.00015660543499332324, "chaser_grad_norm": 0.2612560838460922, "chaser_new_fields": 78.2249984741211, "chaser_partner_vision": 0.0, "chaser_return": 5.575000762939453, "chaser_subspace_norm": 96.18538665771484, "collisions_per_episode": 1.5750000476837158, "entropy": 0.058209532871842384, "explorer_grad_norm": 0.9300043731927872, "explorer_new_fields": 75.20000457763672, "explorer_partner_vision": 0.0, "explorer_return": 62.23749923706055, "explorer_subspace_norm": 26.099992752075195, "final_distance": 5.504971981048584, "max_param_abs": 1.2806016206741333, "policy_loss": -0.0008122388410924941, "value_loss": 3.572008788585663}, "progress_fraction": 0.9, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:30:44.981404+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984644.9814177, "update": 18000}
+{"device": "cuda", "elapsed_hours": 1.9594223218633366, "elapsed_seconds": 7053.920358708012, "env_steps_per_second": 10490.620284456083, "environment_steps": 74000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.15887208015108134, "eta_seconds": 571.9394885438928, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_018000.safetensors", "metrics": {"approx_kl": 0.00017630407430990758, "chaser_grad_norm": 0.1251893099397421, "chaser_new_fields": 79.0250015258789, "chaser_partner_vision": 0.0, "chaser_return": 5.695001125335693, "chaser_subspace_norm": 95.90496063232422, "collisions_per_episode": 1.475000023841858, "entropy": 0.10102851688861847, "explorer_grad_norm": 0.8435127064585686, "explorer_new_fields": 68.20000457763672, "explorer_partner_vision": 0.0, "explorer_return": 51.70000076293945, "explorer_subspace_norm": 26.99217987060547, "final_distance": 4.385324954986572, "max_param_abs": 1.3020002841949463, "policy_loss": -0.0010951019522043026, "value_loss": 3.8845006227493286}, "progress_fraction": 0.925, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:33:51.692801+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782984831.6928124, "update": 18500}
+{"device": "cuda", "elapsed_hours": 2.0117075500569706, "elapsed_seconds": 7242.147180205095, "env_steps_per_second": 10494.125306888296, "environment_steps": 76000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.10587934473984056, "eta_seconds": 381.16564106342605, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_019000.safetensors", "metrics": {"approx_kl": 0.00022960279347063314, "chaser_grad_norm": 0.12079357635229826, "chaser_new_fields": 80.7750015258789, "chaser_partner_vision": 0.0, "chaser_return": 6.040000915527344, "chaser_subspace_norm": 102.02371215820312, "collisions_per_episode": 1.5, "entropy": 0.07814025320112705, "explorer_grad_norm": 1.5280047059059143, "explorer_new_fields": 72.5999984741211, "explorer_partner_vision": 0.0, "explorer_return": 58.29999923706055, "explorer_subspace_norm": 24.30501365661621, "final_distance": 4.763765335083008, "max_param_abs": 1.2802588939666748, "policy_loss": -0.0002921351334101985, "value_loss": 3.4545443654060364}, "progress_fraction": 0.95, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:36:59.919623+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782985019.919637, "update": 19000}
+{"device": "cuda", "elapsed_hours": 2.064875669445286, "elapsed_seconds": 7433.552410003031, "env_steps_per_second": 10492.964291882648, "environment_steps": 78000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.05294552998577656, "eta_seconds": 190.60390794879564, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_019000.safetensors", "metrics": {"approx_kl": 0.0006159323773005099, "chaser_grad_norm": 0.057113331742584705, "chaser_new_fields": 81.625, "chaser_partner_vision": 0.0, "chaser_return": 6.265000820159912, "chaser_subspace_norm": 100.68077850341797, "collisions_per_episode": 0.9000000357627869, "entropy": 0.07017142325639725, "explorer_grad_norm": 1.1960067749023438, "explorer_new_fields": 71.6500015258789, "explorer_partner_vision": 0.0, "explorer_return": 57.10000228881836, "explorer_subspace_norm": 27.36815071105957, "final_distance": 4.832053184509277, "max_param_abs": 1.3260666131973267, "policy_loss": -0.0004417272417356344, "value_loss": 3.56430321931839}, "progress_fraction": 0.975, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:40:11.324852+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782985211.324863, "update": 19500}
+{"device": "cuda", "elapsed_hours": 2.1162682046338968, "elapsed_seconds": 7618.565536682028, "env_steps_per_second": 10500.664411799613, "environment_steps": 80000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.0, "eta_seconds": 0.0, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/update_019000.safetensors", "metrics": {"approx_kl": 0.0007710583850932262, "chaser_grad_norm": 0.11076641827821732, "chaser_new_fields": 83.25, "chaser_partner_vision": 0.0, "chaser_return": 6.595000743865967, "chaser_subspace_norm": 107.82763671875, "collisions_per_episode": 0.9750000238418579, "entropy": 0.06922571919858456, "explorer_grad_norm": 0.3988708183169365, "explorer_new_fields": 73.5, "explorer_partner_vision": 0.0, "explorer_return": 59.76250076293945, "explorer_subspace_norm": 27.074764251708984, "final_distance": 4.443612575531006, "max_param_abs": 1.3260666131973267, "policy_loss": -0.0006433645908252572, "value_loss": 3.4030256271362305}, "progress_fraction": 1.0, "schema_version": 2, "state": "running", "timestamp": "2026-07-02T09:43:16.337978+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782985396.3379903, "update": 20000}
+{"device": "cuda", "elapsed_hours": 2.116271049076135, "elapsed_seconds": 7618.575776674086, "env_steps_per_second": 10500.650298043534, "environment_steps": 80000000, "error": null, "estimated_cost_usd": null, "eta_hours": 0.0, "eta_seconds": 0.0, "healthy": true, "latest_checkpoint": "runs/mouse-run-run-2-mlp/non_social/seed_0001/attempt_01/checkpoints/latest.safetensors", "metrics": {"approx_kl": 0.0007710583850932262, "chaser_grad_norm": 0.11076641827821732, "chaser_new_fields": 83.25, "chaser_partner_vision": 0.0, "chaser_return": 6.595000743865967, "chaser_subspace_norm": 107.82763671875, "collisions_per_episode": 0.9750000238418579, "entropy": 0.06922571919858456, "explorer_grad_norm": 0.3988708183169365, "explorer_new_fields": 73.5, "explorer_partner_vision": 0.0, "explorer_return": 59.76250076293945, "explorer_subspace_norm": 27.074764251708984, "final_distance": 4.443612575531006, "max_param_abs": 1.3260666131973267, "policy_loss": -0.0006433645908252572, "value_loss": 3.4030256271362305}, "progress_fraction": 1.0, "schema_version": 2, "state": "completed", "timestamp": "2026-07-02T09:43:16.348219+00:00", "total_environment_steps": 80000000, "total_updates": 20000, "unix_time": 1782985396.3482296, "update": 20000}