{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":730932260,"defaultBranch":"main","name":"torchtitan","ownerLogin":"pytorch","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-12-13T01:51:37.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/21003710?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1716437563.0","currentOid":""},"activityList":{"items":[{"before":"fadb3abab3c9ec7cc8698ce5184e6c39709e2f5b","after":"94abc6a07052d396fd6d788388f1eb616cb965b2","ref":"refs/heads/gh/wconstab/25/orig","pushedAt":"2024-05-23T04:12:43.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Add 3D support\n\nghstack-source-id: e3e671f6d27d1215baa565463860eef052be18f9\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/344","shortMessageHtmlLink":"Add 3D support"}},{"before":null,"after":"e64dc26a8e8383ddd3f42ad4a4361b7fc87ebac7","ref":"refs/heads/gh/wconstab/27/orig","pushedAt":"2024-05-23T04:12:43.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update pipelining import after change on pytorch\n\nAPIs conform to the pytorch rules.  This PR should be able to land\nsafely after tonight's nightly pytorch build which includes the above\nPR.\n\nghstack-source-id: c575bc7835472128c09798544caa38bf1908e5ca\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/356","shortMessageHtmlLink":"Update pipelining import after change on pytorch"}},{"before":"9884f4247486f070fdb22626e59b3d4d64d4f55d","after":"44ea24c5e01f1ae4ff1af2a06979d0532687f5de","ref":"refs/heads/gh/wconstab/25/head","pushedAt":"2024-05-23T04:12:40.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":"7ede6a30062f787648b5525d60302d669a9aa41a","after":"17b7becc1edf067fc4cc0461de985bf97b021ab3","ref":"refs/heads/gh/wconstab/25/base","pushedAt":"2024-05-23T04:12:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update (base update)\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update (base update)"}},{"before":null,"after":"874379402209a8baaaa76f1935a5e7abbc2810b3","ref":"refs/heads/gh/wconstab/27/head","pushedAt":"2024-05-23T04:12:32.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":null,"after":"979df94778c7d291e6f3e7d697c8719642fcb3db","ref":"refs/heads/gh/wconstab/27/base","pushedAt":"2024-05-23T04:12:32.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update (base update)\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update (base update)"}},{"before":"53b45a0ae9fdd53803be6a0765319fb78073a43b","after":"77cff881bac139e2e1071a08f74e8580ac8bb05e","ref":"refs/heads/gh/tianyu-l/12/orig","pushedAt":"2024-05-22T23:34:09.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"tianyu-l","name":null,"path":"/tianyu-l","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/150487191?s=80&v=4"},"commit":{"message":"only produce tensorboard logs on rank 0 by default\n\nghstack-source-id: d38148fed2e51654b45b59a086cd5bac03e77179\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/339","shortMessageHtmlLink":"only produce tensorboard logs on rank 0 by default"}},{"before":"0cc8c9c0d07a40a7bff3981cbee1f3b016c0ff65","after":"34f85287bc555604a445b348e4b45ab68b4964d8","ref":"refs/heads/gh/tianyu-l/12/head","pushedAt":"2024-05-22T23:34:07.000Z","pushType":"push","commitsCount":6,"pusher":{"login":"tianyu-l","name":null,"path":"/tianyu-l","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/150487191?s=80&v=4"},"commit":{"message":"Update on \"only produce tensorboard logs on rank 0 by default\"\n\n\r\n1. For tensorboard metrics, we mostly care about loss, memory, wps/mfu. Loss is all-reduced so will be the same on all ranks; other metrics are likely to be very similar among all ranks. So by default it suffices to only do tb logging on rank 0 -- the straggler effect should be small for tb writes. User could always toggle on all-rank logging for debugging purposes.\r\n\r\n2. Remove `torch` dependency in `requirements.txt` as it cannot work alone / is not used anyways. Currently we are suggesting users to install latest nightly in README, and do so in all the CI tests.\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update on \"only produce tensorboard logs on rank 0 by default\""}},{"before":"95fa9dd3230d0ba3fd3038d816c4a53dae637462","after":"51f0d4a402d34c3a8792f4794b6d408dc6594310","ref":"refs/heads/gh/tianyu-l/12/base","pushedAt":"2024-05-22T23:34:06.000Z","pushType":"push","commitsCount":5,"pusher":{"login":"tianyu-l","name":null,"path":"/tianyu-l","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/150487191?s=80&v=4"},"commit":{"message":"Update base for Update on \"only produce tensorboard logs on rank 0 by default\"\n\n\r\n1. For tensorboard metrics, we mostly care about loss, memory, wps/mfu. Loss is all-reduced so will be the same on all ranks; other metrics are likely to be very similar among all ranks. So by default it suffices to only do tb logging on rank 0 -- the straggler effect should be small for tb writes. User could always toggle on all-rank logging for debugging purposes.\r\n\r\n2. Remove `torch` dependency in `requirements.txt` as it cannot work alone / is not used anyways. Currently we are suggesting users to install latest nightly in README, and do so in all the CI tests.\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update base for Update on \"only produce tensorboard logs on rank 0 by…"}},{"before":"a1405e856432732da700eb407946dcf1f05fdb93","after":null,"ref":"refs/heads/gh/wconstab/27/orig","pushedAt":"2024-05-22T20:48:18.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"}},{"before":"08aca376b16002ba454b69c9bbf263b66445447c","after":null,"ref":"refs/heads/gh/wconstab/27/head","pushedAt":"2024-05-22T20:48:18.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"}},{"before":"08aca376b16002ba454b69c9bbf263b66445447c","after":null,"ref":"refs/heads/gh/wconstab/27/base","pushedAt":"2024-05-22T20:48:18.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"}},{"before":"6807909215073dab873dbef72ae90f4de1893efe","after":"638ec48941292914075222265fa0d8aadd95df99","ref":"refs/heads/main","pushedAt":"2024-05-22T20:48:15.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Fix bug in PP output layer shape\n\nmostly harmless bug, since output shape of last layer is not used for\nsend/recv purpose, the runtime value overrides it no matter what value\nyou configured it with.\n\nHowever, since adding in/out shape validation to pipeline lib in torch,\nthis raises an error and has to be fixed.\n\nghstack-source-id: 950e41529b7b506085ab280d8a492e345eaefd24\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/354","shortMessageHtmlLink":"Fix bug in PP output layer shape"}},{"before":"52156bb74c1bb8ea6b4ea899f6bfa797842697bc","after":"08aca376b16002ba454b69c9bbf263b66445447c","ref":"refs/heads/gh/wconstab/27/base","pushedAt":"2024-05-22T20:48:13.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":"910662c12060a246e35d4ee72bfa91aa2f642249","after":"6807909215073dab873dbef72ae90f4de1893efe","ref":"refs/heads/main","pushedAt":"2024-05-22T20:10:28.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"wanchaol","name":"Wanchao","path":"/wanchaol","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9443650?s=80&v=4"},"commit":{"message":"Fix llama_13b.toml -> llama2_13b.toml in multinode_trainer.slurm (#350)\n\ncc @wanchaol @lessw2020 @wconstab","shortMessageHtmlLink":"Fix llama_13b.toml -&gt; llama2_13b.toml in multinode_trainer.slurm (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2308918349\" data-permission-text=\"Title is private\" data-url=\"https://github.com/pytorch/torchtitan/issues/350\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/pytorch/torchtitan/pull/350/hovercard\" href=\"https://github.com/pytorch/torchtitan/pull/350\">#350</a>)"}},{"before":"7d8bc0fd81569c0c30e350e66ec213ba4866c356","after":"82faac956ea614263d25cac74d23614358b90036","ref":"refs/heads/gh/wconstab/26/orig","pushedAt":"2024-05-22T16:27:05.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Add a 3-stage PP config\n\nPipelining is unique in that there is no need to stick to power-of-2\nnumbers of stages, and there maybe reasons an odd number is optimal\ndepending on how you divide up your cluster.\n\nAnyway, I use this for validation of the 1f1b schedule in a slightly-more-complicated\nthan 2-stage but simpler than 4-stage setup.\n\nseems to run fine, if run with an even batch size\n(`--training.batch_size 12`)\n\nghstack-source-id: 289eeb8473afa84e3b767986f9fb285f1d91fbf2\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/345","shortMessageHtmlLink":"Add a 3-stage PP config"}},{"before":null,"after":"a1405e856432732da700eb407946dcf1f05fdb93","ref":"refs/heads/gh/wconstab/27/orig","pushedAt":"2024-05-22T16:27:05.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Fix bug in PP output layer shape\n\nmostly harmless bug, since output shape of last layer is not used for\nsend/recv purpose, the runtime value overrides it no matter what value\nyou configured it with.\n\nHowever, since adding in/out shape validation to pipeline lib in torch,\nthis raises an error and has to be fixed.\n\nghstack-source-id: 950e41529b7b506085ab280d8a492e345eaefd24\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/354","shortMessageHtmlLink":"Fix bug in PP output layer shape"}},{"before":"bdba1a1bf36ee4e22861bf4cdb41527a149359b7","after":"fadb3abab3c9ec7cc8698ce5184e6c39709e2f5b","ref":"refs/heads/gh/wconstab/25/orig","pushedAt":"2024-05-22T16:27:05.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Add 3D support\n\nghstack-source-id: fcdac0dd915e229390ceb6745518cedc169d50c2\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/344","shortMessageHtmlLink":"Add 3D support"}},{"before":"f7c999ab010141228cd238befd993626b556262c","after":"822adee4171277afbb5436f83d805d238ccece4d","ref":"refs/heads/gh/wconstab/26/head","pushedAt":"2024-05-22T16:27:02.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":"87171920a87b99ff174b1346dfd10f9fbbcf4a6a","after":"9884f4247486f070fdb22626e59b3d4d64d4f55d","ref":"refs/heads/gh/wconstab/25/head","pushedAt":"2024-05-22T16:27:02.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":"c76bb1587383727caa321824fac46037bcd56db6","after":"496e871d164f744c2d627b24fbebacbe5b6ff5f5","ref":"refs/heads/gh/wconstab/26/base","pushedAt":"2024-05-22T16:27:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update (base update)\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update (base update)"}},{"before":"cca888785ce5e95ae8114a6f4d4c8a7033958006","after":"7ede6a30062f787648b5525d60302d669a9aa41a","ref":"refs/heads/gh/wconstab/25/base","pushedAt":"2024-05-22T16:27:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update (base update)\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update (base update)"}},{"before":null,"after":"08aca376b16002ba454b69c9bbf263b66445447c","ref":"refs/heads/gh/wconstab/27/head","pushedAt":"2024-05-22T16:26:53.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":null,"after":"52156bb74c1bb8ea6b4ea899f6bfa797842697bc","ref":"refs/heads/gh/wconstab/27/base","pushedAt":"2024-05-22T16:26:53.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update (base update)\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update (base update)"}},{"before":"a5bfd9cd549604bcc2176fd64a14c2b526d0271f","after":null,"ref":"refs/heads/gh/wconstab/27/orig","pushedAt":"2024-05-22T16:24:45.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"}},{"before":"3f25553b85ef1bafe093ad156fce19ce5643cc48","after":null,"ref":"refs/heads/gh/wconstab/27/head","pushedAt":"2024-05-22T16:24:45.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"}},{"before":"3f25553b85ef1bafe093ad156fce19ce5643cc48","after":null,"ref":"refs/heads/gh/wconstab/27/base","pushedAt":"2024-05-22T16:24:45.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"}},{"before":"60810a9e7620a554b52ce5d52ed16c8bc7fe56be","after":"910662c12060a246e35d4ee72bfa91aa2f642249","ref":"refs/heads/main","pushedAt":"2024-05-22T16:24:43.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Make test_runner use separate logger with default INFO\n\nprevious change to use logging from torchtitan caused stdout not\nto show up.\n\nghstack-source-id: 30a77c59ba68043ffa844be0443d5351d9584fab\nPull Request resolved: https://github.com/pytorch/torchtitan/pull/352","shortMessageHtmlLink":"Make test_runner use separate logger with default INFO"}},{"before":"884a656e207f2c072cea06fad52358a52fcd863e","after":"3f25553b85ef1bafe093ad156fce19ce5643cc48","ref":"refs/heads/gh/wconstab/27/base","pushedAt":"2024-05-22T16:24:40.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"wconstab","name":"Will Constable","path":"/wconstab","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/4984825?s=80&v=4"},"commit":{"message":"Update\n\n[ghstack-poisoned]","shortMessageHtmlLink":"Update"}},{"before":"81dc9e37d5d919f454c44e11b9216be46702c63f","after":null,"ref":"refs/heads/compile_2d","pushedAt":"2024-05-22T05:10:07.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"wanchaol","name":"Wanchao","path":"/wanchaol","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/9443650?s=80&v=4"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEUZX22AA","startCursor":null,"endCursor":null}},"title":"Activity · pytorch/torchtitan"}