{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":795075392,"defaultBranch":"main","name":"self-reasoning-tokens-pytorch","ownerLogin":"lucidrains","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2024-05-02T14:33:47.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/108653?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1714932063.0","currentOid":""},"activityList":{"items":[{"before":"e344e831bd047c41a23dc0f24381ca19c1ce58f7","after":"cfc93ec82d05ba462ab0022cda152d8f5b0abc29","ref":"refs/heads/main","pushedAt":"2024-05-17T01:51:48.000Z","pushType":"pr_merge","commitsCount":2,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"Merge pull request #1 from FeSens/main\n\ndocs(README.md): It was actually this year ;)","shortMessageHtmlLink":"Merge pull request #1 from FeSens/main"}},{"before":"5e0cef059b85d5bc03c7cf36a93e796df5fd8afa","after":"e344e831bd047c41a23dc0f24381ca19c1ce58f7","ref":"refs/heads/main","pushedAt":"2024-05-05T18:01:00.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"allow also for scaling the grads near and far for queries, keys, values","shortMessageHtmlLink":"allow also for scaling the grads near and far for queries, keys, values"}},{"before":"edc373f33498922572a5e779fe8bf9841ad355e2","after":"5e0cef059b85d5bc03c7cf36a93e796df5fd8afa","ref":"refs/heads/main","pushedAt":"2024-05-05T18:00:19.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"allow also for scaling the grads near and far for queries, keys, values","shortMessageHtmlLink":"allow also for scaling the grads near and far for queries, keys, values"}},{"before":"3a346d3235d4b4fc1f3fc625f88ee826cb0a0a64","after":"edc373f33498922572a5e779fe8bf9841ad355e2","ref":"refs/heads/main","pushedAt":"2024-05-03T18:46:43.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"readme","shortMessageHtmlLink":"readme"}},{"before":"9f5851c27a4ac53a2599b6ad0277fba9b6caf552","after":"3a346d3235d4b4fc1f3fc625f88ee826cb0a0a64","ref":"refs/heads/main","pushedAt":"2024-05-03T18:43:12.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"readme","shortMessageHtmlLink":"readme"}},{"before":"e781ea3b31fcd34cf4d068df76b8a90f7aa14eac","after":"9f5851c27a4ac53a2599b6ad0277fba9b6caf552","ref":"refs/heads/main","pushedAt":"2024-05-03T18:38:41.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"readme","shortMessageHtmlLink":"readme"}},{"before":"95502af9e81f14314a7c79bde7e60f435c597872","after":"e781ea3b31fcd34cf4d068df76b8a90f7aa14eac","ref":"refs/heads/main","pushedAt":"2024-05-03T14:53:06.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"export full multi head causal attention with stop graddable qkv for experiments","shortMessageHtmlLink":"export full multi head causal attention with stop graddable qkv for e…"}},{"before":"3c6a5e77b59fef5be5124bfb23704d5da68436f1","after":"95502af9e81f14314a7c79bde7e60f435c597872","ref":"refs/heads/main","pushedAt":"2024-05-03T14:37:02.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"offer a naive unoptimized attention w/ stop graddable queries, keys, values, and fix the self reasoning transformer to only stop grad keys and values for the reasoning tokens","shortMessageHtmlLink":"offer a naive unoptimized attention w/ stop graddable queries, keys, …"}},{"before":"314889206e6baa0378d4d201f91e2e3cb16d6704","after":"3c6a5e77b59fef5be5124bfb23704d5da68436f1","ref":"refs/heads/main","pushedAt":"2024-05-02T20:19:51.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"allow for returning logits or calculating loss without reasoning tokens","shortMessageHtmlLink":"allow for returning logits or calculating loss without reasoning tokens"}},{"before":"6e8894204be84709c66045738d0169c410f137d3","after":"314889206e6baa0378d4d201f91e2e3cb16d6704","ref":"refs/heads/main","pushedAt":"2024-05-02T19:33:52.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"will try original idea, where next token gradients are stop gradded properly for queries and keys, but not values","shortMessageHtmlLink":"will try original idea, where next token gradients are stop gradded p…"}},{"before":"406afd174bf8a98792a6e55a910d203ea5316816","after":"6e8894204be84709c66045738d0169c410f137d3","ref":"refs/heads/main","pushedAt":"2024-05-02T19:23:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"fix mask","shortMessageHtmlLink":"fix mask"}},{"before":"9073488ef32aaee405dcbae86bcd95714c47d11b","after":"406afd174bf8a98792a6e55a910d203ea5316816","ref":"refs/heads/main","pushedAt":"2024-05-02T19:11:11.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"start with straightforward attention masking of reasoning tokens","shortMessageHtmlLink":"start with straightforward attention masking of reasoning tokens"}},{"before":"13d8c92242d9b1ed329ee123b036d4d5c267beca","after":"9073488ef32aaee405dcbae86bcd95714c47d11b","ref":"refs/heads/main","pushedAt":"2024-05-02T14:59:43.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"readme","shortMessageHtmlLink":"readme"}},{"before":"5d2e86e5b628477aed553b4977fbde9eb65840ba","after":"13d8c92242d9b1ed329ee123b036d4d5c267beca","ref":"refs/heads/main","pushedAt":"2024-05-02T14:58:13.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"add the most telling diagram from the blog post","shortMessageHtmlLink":"add the most telling diagram from the blog post"}},{"before":"efca355eec340980da960000be54c8c2c63aa7fb","after":"5d2e86e5b628477aed553b4977fbde9eb65840ba","ref":"refs/heads/main","pushedAt":"2024-05-02T14:56:41.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"add the most telling diagram from the blog post","shortMessageHtmlLink":"add the most telling diagram from the blog post"}},{"before":"d98de1b9d9d504c67a6770ce148247ea37b5504b","after":"efca355eec340980da960000be54c8c2c63aa7fb","ref":"refs/heads/main","pushedAt":"2024-05-02T14:54:10.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"readme","shortMessageHtmlLink":"readme"}},{"before":null,"after":"d98de1b9d9d504c67a6770ce148247ea37b5504b","ref":"refs/heads/main","pushedAt":"2024-05-02T14:33:48.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"lucidrains","name":"Phil Wang","path":"/lucidrains","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/108653?s=80&v=4"},"commit":{"message":"Initial commit","shortMessageHtmlLink":"Initial commit"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAETHKaoAA","startCursor":null,"endCursor":null}},"title":"Activity · lucidrains/self-reasoning-tokens-pytorch"}