Commit Graph

931 Commits

Author SHA1 Message Date
henk717 f1b0ea711e
Merge branch 'KoboldAI:main' into united 2022-03-06 19:02:59 +01:00
henk717 932aabc2f3
Merge pull request #103 from henk717/main
Modern ROCm Docker
2022-03-06 19:02:38 +01:00
henk717 4332074c89 Modern ROCm Docker
Brings the ROCm container up to a modern standard in line with the CUDA docker.
2022-03-06 19:01:25 +01:00
henk717 4835192041 Load TK on demand 2022-03-06 14:12:01 +01:00
henk717 daea4b8d15 Fix Breakmodel RAM Regression 2022-03-06 08:26:50 +01:00
henk717 105d3831b5 Lazy Load Float32 for CPU 2022-03-06 07:56:04 +01:00
henk717 77cc2ee789
Merge pull request #93 from VE-FORBRYDERNE/lazy-loader
Lazy loader
2022-03-05 20:32:31 +01:00
Gnome Ann 373f7b9bd5 Don't convert tensors to float16 if using CPU-only mode 2022-03-05 14:30:26 -05:00
Gnome Ann 579e85820c Resolve merge conflict 2022-03-05 14:13:56 -05:00
Gnome Ann 2e19ea1bb6 Auto detect if we're in a Colab TPU instance 2022-03-05 14:07:23 -05:00
henk717 3a5793c815 No longer uses --colab_tpu 2022-03-05 19:58:24 +01:00
henk717 935c7e5786 Improved TPU support 2022-03-05 19:47:51 +01:00
henk717 6f2febb142
Merge pull request #92 from ebolam/united
Hopefully Last Redo Fix
2022-03-05 19:26:15 +01:00
ebolam 4a8d7f5e0b
Merge branch 'henk717:united' into united 2022-03-05 13:25:10 -05:00
henk717 c20435855b
Merge pull request #91 from VE-FORBRYDERNE/transformers-version-check
Put the XGLM embedding patch behind a version check
2022-03-05 19:03:00 +01:00
Gnome Ann 4625158d30 Fix typo in previous commit 2022-03-05 12:56:42 -05:00
Gnome Ann 0a258a6282 Support for loading HF models on TPU with `--colab_tpu` 2022-03-05 12:33:33 -05:00
Gnome Ann 86ac562b0c Lazy loader should convert model tensors to float16 before moving them 2022-03-05 11:31:34 -05:00
ebolam 4dd119c38d Redo no longer goes through formatting function (thereby getting changed) 2022-03-05 11:15:33 -05:00
ebolam 353817b4da Remove debug print statements 2022-03-05 10:35:06 -05:00
ebolam 221f264fa7 Redo fix. Fix for actions structure to not error out when asking for next_id when the actions list is empty. 2022-03-05 10:31:28 -05:00
Gnome Ann a00dede610 Put the XGLM embedding patch behind a version check 2022-03-04 19:10:15 -05:00
Gnome Ann 5674516f0c Merge branch 'united' into lazy-loader 2022-03-04 18:27:51 -05:00
henk717 8e12b7df61
Merge pull request #90 from ebolam/united
Redo Bug Fix
2022-03-04 22:10:49 +01:00
ebolam 5f92cbc231 Merge branch 'united' of https://github.com/ebolam/KoboldAI into united 2022-03-04 15:37:34 -05:00
ebolam 321f45ccad Fix debug to never crash (would on some initialization steps) 2022-03-04 15:36:13 -05:00
ebolam ee883fc4da
Merge branch 'henk717:united' into united 2022-03-04 14:15:16 -05:00
ebolam 26b9268391 Redo bug fix 2022-03-04 14:14:44 -05:00
henk717 eb247d69c3
Merge branch 'KoboldAI:main' into united 2022-03-04 18:24:56 +01:00
henk717 657de72ada
Merge: Better name formatting for chatmode 2022-03-04 18:24:39 +01:00
Gnome Ann 4474607f88 Merge branch 'united' into lazy-loader 2022-03-04 11:12:29 -05:00
Gnome Ann a1fedca2c8 Use lazy loading automatically if a config file exists for the model 2022-03-04 11:11:33 -05:00
MrReplikant ff1be78f72
Merge pull request #1 from MrReplikant/MrReplikant-patch-1
Fixed unnecessary spacing in chatmode
2022-03-04 08:46:43 -06:00
MrReplikant ae143e896c
Fixed unnecessary spacing in chatmode
This makes it go from "john :" to "John:", as it's supposed to be. As simple as it is, it can easily throw a chatbot model for a loop.
2022-03-04 08:46:00 -06:00
henk717 addc7edd49
Merge branch 'KoboldAI:main' into united 2022-03-04 11:34:04 +01:00
henk717 749d4a1c48 Update Colab Descriptions (GPU) 2022-03-04 11:33:05 +01:00
henk717 fade5fdd60 Update model descriptions (TPU) 2022-03-04 11:31:03 +01:00
henk717 2936778dbc
Merge branch 'KoboldAI:main' into united 2022-03-04 09:56:35 +01:00
henk717 2aeb2c6607 Add Janeway 6B and Shinen 6B 2022-03-04 09:53:34 +01:00
Gnome Ann f0629958b1 Merge branch 'united' into lazy-loader 2022-03-04 00:37:25 -05:00
Gnome Ann 58a2c18821 Add lazy torch loading support to transformers backend 2022-03-04 00:33:10 -05:00
Gnome Ann 1515996fca Fix torch_lazy_loader seek offset calculation 2022-03-03 23:53:40 -05:00
Gnome Ann 24bc0f81ea Remove duplicate `torch_load` definition 2022-03-03 19:55:31 -05:00
Gnome Ann 8e6e04be5f (torch_lazy_loader.py) Add dematerialized modules setting 2022-03-03 11:17:59 -05:00
Gnome Ann 1ecc452dc8 (torch_lazy_loader.py) Add support for materializing from a ZipExtFile 2022-03-02 13:08:21 -05:00
henk717 e033b04f87 Restore United 2022-03-02 11:40:50 +01:00
henk717 f9ac23ba4e Add Janeway and Shinen 2022-03-02 09:51:25 +01:00
henk717 c8ece04b1d
Merge pull request #99 from VE-FORBRYDERNE/mutation-observer
Re-enable the editor mutation observer
2022-03-02 09:39:03 +01:00
Gnome Ann c338b52d68 (torch_lazy_loader.py) Handle checkpoints with merged storage blocks 2022-03-02 01:02:35 -05:00
Gnome Ann 4fa4dbac50 Clean up when error is thrown in `use_lazy_torch_load` 2022-03-01 19:30:22 -05:00