{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":653798952,"defaultBranch":"master","name":"NEKO","ownerLogin":"ManifoldRG","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-06-14T19:00:22.000Z","ownerAvatar":"https://github.com/avatars/u/64051512?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1718839576.0","currentOid":""},"activityList":{"items":[{"before":null,"after":"075d72d8eec9235357ab722e31998f67b6b8d4b3","ref":"refs/heads/imprv_loss","pushedAt":"2024-06-19T23:26:16.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Merge pull request #85 from ManifoldRG/fix_Accelerator_deprecation\n\nFix the deprecation of the usage Accelerator, add code for new way","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2216591475\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ManifoldRG/NEKO/issues/85\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ManifoldRG/NEKO/pull/85/hovercard\" href=\"https://github.com/ManifoldRG/NEKO/pull/85\">#85</a> from ManifoldRG/fix_Accelerator_deprecation"}},{"before":null,"after":"fdcca1b76352f305850cc0d3c0cb8987a7a5b416","ref":"refs/heads/text_fix_integration","pushedAt":"2024-06-06T22:05:20.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"filter empty strings from text dataset\n\nWhen we added `padding='longest'` to the tokenization, it caused empty\nstrings to be padded to the same length as the longest sample. That\nmeant that `numel()` was never going to be 0.\n\nThis commit moves the empty string check to the part of the code that\nloads the dataset into memory.\n\nBecause of the way we're sampling indices in `sample_batch`...:\n\n```\n        sampled_indices = torch.randperm(len(dataset_split))[:batch_size]\n        samples = dataset_split.select(sampled_indices)\n        tokenized_outputs = self.text_tokenizer(samples['text'], truncation=True, padding=\"longest\", max_length=self.context_length, return_tensors='pt')\n```\n\n... we were ending up with batches of varying sizes, depending on how\nmany empty text strings we sampled. We always sample 64 indices, for\nexample, but sometimes we might get 20 of them that are empty strings.\nOther times we might get 10 of them that are empty strings. Therefore\nour batch size would vary on each sample. By filtering when we load the\ndataset into memeory, we guarantee every batch size to be the same.","shortMessageHtmlLink":"filter empty strings from text dataset"}},{"before":"075d72d8eec9235357ab722e31998f67b6b8d4b3","after":null,"ref":"refs/heads/text_fix_integration","pushedAt":"2024-06-06T22:04:30.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"}},{"before":null,"after":"075d72d8eec9235357ab722e31998f67b6b8d4b3","ref":"refs/heads/text_fix_integration","pushedAt":"2024-06-06T22:00:16.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Merge pull request #85 from ManifoldRG/fix_Accelerator_deprecation\n\nFix the deprecation of the usage Accelerator, add code for new way","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2216591475\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ManifoldRG/NEKO/issues/85\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ManifoldRG/NEKO/pull/85/hovercard\" href=\"https://github.com/ManifoldRG/NEKO/pull/85\">#85</a> from ManifoldRG/fix_Accelerator_deprecation"}},{"before":"d85ca82b54cf0e702e0ce0f8a02b694bb5ae6fac","after":"fdcca1b76352f305850cc0d3c0cb8987a7a5b416","ref":"refs/heads/text_fix","pushedAt":"2024-06-03T17:34:47.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"eihli","name":"Eric Ihli","path":"/eihli","primaryAvatarUrl":"https://github.com/avatars/u/1719584?s=80&v=4"},"commit":{"message":"filter empty strings from text dataset\n\nWhen we added `padding='longest'` to the tokenization, it caused empty\nstrings to be padded to the same length as the longest sample. That\nmeant that `numel()` was never going to be 0.\n\nThis commit moves the empty string check to the part of the code that\nloads the dataset into memory.\n\nBecause of the way we're sampling indices in `sample_batch`...:\n\n```\n        sampled_indices = torch.randperm(len(dataset_split))[:batch_size]\n        samples = dataset_split.select(sampled_indices)\n        tokenized_outputs = self.text_tokenizer(samples['text'], truncation=True, padding=\"longest\", max_length=self.context_length, return_tensors='pt')\n```\n\n... we were ending up with batches of varying sizes, depending on how\nmany empty text strings we sampled. We always sample 64 indices, for\nexample, but sometimes we might get 20 of them that are empty strings.\nOther times we might get 10 of them that are empty strings. Therefore\nour batch size would vary on each sample. By filtering when we load the\ndataset into memeory, we guarantee every batch size to be the same.","shortMessageHtmlLink":"filter empty strings from text dataset"}},{"before":"0ab561b6d686bd2560e803144bb7a42fb9605da0","after":"d85ca82b54cf0e702e0ce0f8a02b694bb5ae6fac","ref":"refs/heads/text_fix","pushedAt":"2024-06-03T17:33:49.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"eihli","name":"Eric Ihli","path":"/eihli","primaryAvatarUrl":"https://github.com/avatars/u/1719584?s=80&v=4"},"commit":{"message":"filter empty strings from text dataset\n\nWhen we added `padding='longest'` to the tokenization, it caused empty\nstrings to be padded to the same length as the longest sample. That\nmeant that `numel()` was never going to be 0.\n\nThis commit moves the empty string check to the part of the code that\nloads the dataset into memory.\n\nBecause of the way we're sampling indices in `sample_batch`...:\n\n```\n        sampled_indices = torch.randperm(len(dataset_split))[:batch_size]\n        samples = dataset_split.select(sampled_indices)\n        tokenized_outputs = self.text_tokenizer(samples['text'], truncation=True, padding=\"longest\", max_length=self.context_length, return_tensors='pt')\n```\n\n... we were ending up with batches of varying sizes, depending on how\nmany empty text strings we sampled. We always sample 64 indices, for\nexample, but sometimes we might get 20 of them that are empty strings.\nOther times we might get 10 of them that are empty strings. Therefore\nour batch size would vary on each sample. By filtering when we load the\ndataset into memeory, we guarantee every batch size to be the same.","shortMessageHtmlLink":"filter empty strings from text dataset"}},{"before":"a85ad05be85b05df48ab3fe608aa624342d39ceb","after":"0ab561b6d686bd2560e803144bb7a42fb9605da0","ref":"refs/heads/text_fix","pushedAt":"2024-06-03T17:31:52.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"eihli","name":"Eric Ihli","path":"/eihli","primaryAvatarUrl":"https://github.com/avatars/u/1719584?s=80&v=4"},"commit":{"message":"filter empty strings from text dataset\n\nWhen we added `padding='longest'` to the tokenization, it caused\nempty strings to be padded to the same length as the longest sample.\nThat meant that `numel()` was never going to be 0.\n\nThis commit moves the empty string check to the part of the code that\nloads the dataset into memory.\n\nBecause of the way we're sampling indices in `sample_batch`...:\n\n```\n        sampled_indices = torch.randperm(len(dataset_split))[:batch_size]\n        samples = dataset_split.select(sampled_indices)\n        tokenized_outputs = self.text_tokenizer(samples['text'], truncation=True, padding=\"longest\", max_length=self.context_length, return_tensors='pt')\n```\n\n... we end up with batches of varying sizes, depending on how many empty\ntext strings we sampled. We always sample 64 indices, for example, but\nsometimes we might get 20 of them that are empty strings. Other times\nwe might get 10 of them that are empty strings. Therefore our batch size\nwould very on each sample. By filtering when we load the dataset into\nmemeory, we guarantee every batch size to be the same.","shortMessageHtmlLink":"filter empty strings from text dataset"}},{"before":"368c497638332d5fca56165af2aebf5cd07ee279","after":"bb7d42b9cf58c3931201f9eecb4cdf40e4f1f993","ref":"refs/heads/try_timm_vit","pushedAt":"2024-05-29T03:26:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Test another method to extract image embeddings from timm vit","shortMessageHtmlLink":"Test another method to extract image embeddings from timm vit"}},{"before":"075d72d8eec9235357ab722e31998f67b6b8d4b3","after":"368c497638332d5fca56165af2aebf5cd07ee279","ref":"refs/heads/try_timm_vit","pushedAt":"2024-05-28T11:36:29.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Test the codebase with timm vit","shortMessageHtmlLink":"Test the codebase with timm vit"}},{"before":"5bf0905097e67beaa7526c8b26495d8fa8879572","after":"a85ad05be85b05df48ab3fe608aa624342d39ceb","ref":"refs/heads/text_fix","pushedAt":"2024-05-27T23:25:46.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"fix copying weights for only text token part when using pretrained_lm","shortMessageHtmlLink":"fix copying weights for only text token part when using pretrained_lm"}},{"before":"de437ed9c68435f759da2fa8a470f394b63f204e","after":"5bf0905097e67beaa7526c8b26495d8fa8879572","ref":"refs/heads/text_fix","pushedAt":"2024-05-26T17:20:21.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"re-integrate text-modality fixes to neko codebase - part1/2","shortMessageHtmlLink":"re-integrate text-modality fixes to neko codebase - part1/2"}},{"before":null,"after":"075d72d8eec9235357ab722e31998f67b6b8d4b3","ref":"refs/heads/try_timm_vit","pushedAt":"2024-05-25T12:39:17.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Merge pull request #85 from ManifoldRG/fix_Accelerator_deprecation\n\nFix the deprecation of the usage Accelerator, add code for new way","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2216591475\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ManifoldRG/NEKO/issues/85\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ManifoldRG/NEKO/pull/85/hovercard\" href=\"https://github.com/ManifoldRG/NEKO/pull/85\">#85</a> from ManifoldRG/fix_Accelerator_deprecation"}},{"before":"13fe707e05b88d9bb6c94457a28e308185cb8881","after":"de437ed9c68435f759da2fa8a470f394b63f204e","ref":"refs/heads/text_fix","pushedAt":"2024-05-13T11:59:44.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"add evaluation files","shortMessageHtmlLink":"add evaluation files"}},{"before":"8a1d0a4c99b598f7b323f9cc776e5be06128e0c7","after":"13fe707e05b88d9bb6c94457a28e308185cb8881","ref":"refs/heads/text_fix","pushedAt":"2024-05-13T02:33:03.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"working config for minipile dataset also","shortMessageHtmlLink":"working config for minipile dataset also"}},{"before":"29582249f9fc9724079901c0b0c3c0e82fcdbfac","after":"8a1d0a4c99b598f7b323f9cc776e5be06128e0c7","ref":"refs/heads/text_fix","pushedAt":"2024-05-13T01:58:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"this version works","shortMessageHtmlLink":"this version works"}},{"before":"01f7b7d1f5d89f4c9ab0688e236f8b819469ce5c","after":"29582249f9fc9724079901c0b0c3c0e82fcdbfac","ref":"refs/heads/text_fix","pushedAt":"2024-05-13T00:55:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"add script which can use multiple GPUs","shortMessageHtmlLink":"add script which can use multiple GPUs"}},{"before":"44e1347c35691937b825239d8687207d1d8cf4ed","after":"01f7b7d1f5d89f4c9ab0688e236f8b819469ce5c","ref":"refs/heads/text_fix","pushedAt":"2024-05-11T19:55:32.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"add nucleus sampling back to notebook","shortMessageHtmlLink":"add nucleus sampling back to notebook"}},{"before":"f8f6ce3e7501d4dd73216f285a9eefc88872f96d","after":"44e1347c35691937b825239d8687207d1d8cf4ed","ref":"refs/heads/text_fix","pushedAt":"2024-05-11T07:02:02.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"Fix predict_text for any text generation","shortMessageHtmlLink":"Fix predict_text for any text generation"}},{"before":"003d4ae68bb89e52fb4630b5db6d3aa5a0535be0","after":"f8f6ce3e7501d4dd73216f285a9eefc88872f96d","ref":"refs/heads/text_fix","pushedAt":"2024-05-11T06:23:42.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"make sure pretrained flag works too","shortMessageHtmlLink":"make sure pretrained flag works too"}},{"before":"e7e327080f72a41ee188b83fd2ee99cc03cfdf80","after":"003d4ae68bb89e52fb4630b5db6d3aa5a0535be0","ref":"refs/heads/text_fix","pushedAt":"2024-05-11T05:48:03.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"perplexity is fine now","shortMessageHtmlLink":"perplexity is fine now"}},{"before":"53a88bd7b25ec657c3859ef6aaa5c38d885b41a5","after":"e7e327080f72a41ee188b83fd2ee99cc03cfdf80","ref":"refs/heads/text_fix","pushedAt":"2024-05-11T04:07:51.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"add a couple of notebooks that are easier to work with at least","shortMessageHtmlLink":"add a couple of notebooks that are easier to work with at least"}},{"before":null,"after":"53a88bd7b25ec657c3859ef6aaa5c38d885b41a5","ref":"refs/heads/text_fix","pushedAt":"2024-05-08T19:10:31.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"test nb","shortMessageHtmlLink":"test nb"}},{"before":"71d1a9d98bd035408492a6e826005be40bee986a","after":"075d72d8eec9235357ab722e31998f67b6b8d4b3","ref":"refs/heads/master","pushedAt":"2024-05-07T23:33:29.000Z","pushType":"pr_merge","commitsCount":4,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Merge pull request #85 from ManifoldRG/fix_Accelerator_deprecation\n\nFix the deprecation of the usage Accelerator, add code for new way","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2216591475\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ManifoldRG/NEKO/issues/85\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ManifoldRG/NEKO/pull/85/hovercard\" href=\"https://github.com/ManifoldRG/NEKO/pull/85\">#85</a> from ManifoldRG/fix_Accelerator_deprecation"}},{"before":"aa3239ccdf8f9851d966f4afadeac5e1b3a3dd85","after":"2c9d9f38d839de7c187ac9bb8c84b50728393108","ref":"refs/heads/fix_Accelerator_deprecation","pushedAt":"2024-05-07T23:11:49.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"update accelerate version","shortMessageHtmlLink":"update accelerate version"}},{"before":"d1f6526eb95c962f1677c1fa8db22a37e6fd2a51","after":"aa3239ccdf8f9851d966f4afadeac5e1b3a3dd85","ref":"refs/heads/fix_Accelerator_deprecation","pushedAt":"2024-05-07T21:47:47.000Z","pushType":"push","commitsCount":10,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Merge branch 'master' into fix_Accelerator_deprecation","shortMessageHtmlLink":"Merge branch 'master' into fix_Accelerator_deprecation"}},{"before":"71d1a9d98bd035408492a6e826005be40bee986a","after":"220a7bdeab6f5355432ad2b6a74c505125788d2f","ref":"refs/heads/test_ViT","pushedAt":"2024-05-05T11:38:42.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Test various ViT in NEKO","shortMessageHtmlLink":"Test various ViT in NEKO"}},{"before":null,"after":"71d1a9d98bd035408492a6e826005be40bee986a","ref":"refs/heads/test_ViT","pushedAt":"2024-05-05T10:35:32.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"Merge pull request #80 from eihli/distributed-wandb\n\nFix issues with distributed training","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2131334634\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ManifoldRG/NEKO/issues/80\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ManifoldRG/NEKO/pull/80/hovercard\" href=\"https://github.com/ManifoldRG/NEKO/pull/80\">#80</a> from eihli/distributed-wandb"}},{"before":"0a257c38b92be39bfcf31889fafd8373954550e3","after":"db4756bf35ac9c479624bfee02d7aae95e11faf6","ref":"refs/heads/fix_VQA_loss","pushedAt":"2024-05-05T10:23:31.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"This is a test to see whether it can reduce VQA loss","shortMessageHtmlLink":"This is a test to see whether it can reduce VQA loss"}},{"before":"db4756bf35ac9c479624bfee02d7aae95e11faf6","after":"0a257c38b92be39bfcf31889fafd8373954550e3","ref":"refs/heads/fix_VQA_loss","pushedAt":"2024-05-05T02:57:25.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"henryj18","name":null,"path":"/henryj18","primaryAvatarUrl":"https://github.com/avatars/u/49120145?s=80&v=4"},"commit":{"message":"This version tests replacing the ImageEmbedding with LucidRains ViT","shortMessageHtmlLink":"This version tests replacing the ImageEmbedding with LucidRains ViT"}},{"before":"0bf709e482b74351efcbef35b00f4f3f548c0c38","after":"71d1a9d98bd035408492a6e826005be40bee986a","ref":"refs/heads/master","pushedAt":"2024-04-20T18:23:20.000Z","pushType":"pr_merge","commitsCount":9,"pusher":{"login":"bhavul","name":"Bhavul Gauri","path":"/bhavul","primaryAvatarUrl":"https://github.com/avatars/u/5140951?s=80&v=4"},"commit":{"message":"Merge pull request #80 from eihli/distributed-wandb\n\nFix issues with distributed training","shortMessageHtmlLink":"Merge pull request <a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2131334634\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ManifoldRG/NEKO/issues/80\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ManifoldRG/NEKO/pull/80/hovercard\" href=\"https://github.com/ManifoldRG/NEKO/pull/80\">#80</a> from eihli/distributed-wandb"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEaf_qZQA","startCursor":null,"endCursor":null}},"title":"Activity · ManifoldRG/NEKO"}