rixty/KoboldAI-Client
1
1
Fork 0
mirror of https://github.com/KoboldAI/KoboldAI-Client.git synced 2025-06-05 21:59:24 +02:00
Code Issues Projects Releases Wiki Activity
4,046 Commits 1 Branch 5 Tags
839d56ebf2e7409705a109722bf55edd0fcee77c
Commit Graph

95 Commits

Author SHA1 Message Date
ebolam
839d56ebf2 Potential fix for gpt-neo and gpt-j 2023-05-23 19:25:01 -04:00
ebolam
5561cc1f22 Fix for GPU generation 2023-05-23 08:33:19 -04:00
ebolam
4c25d6fbbb Fix for loading model multiple times loosing the gpu/cpu splits 2023-05-22 20:34:01 -04:00
ebolam
9e53bcf676 Fix for breakmodel loading to CPU when set to GPU 2023-05-22 20:24:57 -04:00
ebolam
f1a16f260f Potential breakmodel fix 2023-05-22 16:10:41 -04:00
ebolam
ca770844b0 Fix for breakmodel 2023-05-22 15:07:59 -04:00
ebolam
dc20e6dde9 Fix for unloading models 2023-05-22 15:04:33 -04:00
ebolam
925cad2e2f Better compatibility with hf model backend 2023-05-22 14:50:13 -04:00
ebolam
513b8575e7 Fix for missing import 
Fix for model name being a path which caused save issues
 2023-05-20 11:01:49 -04:00
ebolam
a1ee6849dc Custom Paths from Menu structure fixed 2023-05-19 18:28:47 -04:00
ebolam
6df5fe4ad0 partial load model from custom path in menu 2023-05-19 18:24:06 -04:00
ebolam
9df1f03b12 Fix for custom huggingface model menu entry 2023-05-19 14:28:36 -04:00
ebolam
36e679b366 Merge branch 'Model_Plugins' of https://github.com/ebolam/KoboldAI into Model_Plugins 2023-05-19 09:11:22 -04:00
ebolam
99cffd4755 Colab GPU edition fixes 2023-05-19 09:11:08 -04:00
ebolam
3db231562f Merge pull request #382 from henk717/united 
Update to united
 2023-05-19 06:05:25 -04:00
ebolam
56d2705f4b removed breakmodel command line arguments (except nobreakmodel) 2023-05-18 20:19:33 -04:00
ebolam
06f59a7b7b Moved model backends to separate folders 
added some model backend settings save/load
 2023-05-18 20:14:33 -04:00
ebolam
4040538d34 Model Backends now defined in the menu 2023-05-18 18:34:00 -04:00
ebolam
182ecff202 Added in model backend to the command line arguments 2023-05-18 16:01:17 -04:00
ebolam
f027d8b6e5 Better working valid detection and named model backends for UI 2023-05-17 21:15:31 -04:00
Henk
205c64f1ea More universal pytorch folder detection 2023-05-13 20:26:55 +02:00
ebolam
c6b17889d0 Updated to latest united 2023-05-12 07:53:27 -04:00
ebolam
aaa9133899 Disk Cache working 
UI valid marker broken for disk cache
 2023-05-11 21:22:33 -04:00
ebolam
a6f0e97ba0 Working(?) breakmodel 2023-05-11 20:40:05 -04:00
ebolam
69d942c00c Kind of working breakmodel 2023-05-11 20:22:30 -04:00
somebody
3065c1b40e Ignore missing keys in get_original_key 2023-05-11 17:10:43 -05:00
somebody
c16336f646 Add traceback to debug log on fallback 2023-05-11 17:10:19 -05:00
ebolam
a9c785d0f0 Fix for Horde 2023-05-11 14:20:14 -04:00
ebolam
e9c845dc2a Fix for badwordIDs 2023-05-11 14:14:52 -04:00
ebolam
4605d10c37 Next iteration. Model Loading is broken completely now :) 2023-05-11 12:08:35 -04:00
ebolam
77dd5aa725 Minor update 2023-05-11 09:09:09 -04:00
Henk
e932364a1e RWKV support 2023-05-11 14:56:12 +02:00
ebolam
71aee4dbd8 First concept of model plugins with a conceptual UI. 
Completely breaks UI2 model loading.
 2023-05-10 16:30:46 -04:00
Bogdan Drema
d53726bed6 fix: tpu tokenizers errors 2023-05-08 18:24:34 +01:00
Henk
bb206f598e Don't load peft when unused 2023-05-06 18:55:26 +02:00
somebody
b7db709c47 PEFT: Change directory structure to be inside model 2023-05-06 11:16:09 -05:00
somebody
f02ddab7c7 Merge branch 'united' of https://github.com/henk717/KoboldAI into peft 2023-05-06 10:47:14 -05:00
Henk
33969b5845 Basic HF code execution support 2023-05-05 17:23:01 +02:00
somebody
35b56117e6 Basic PEFT support 2023-05-03 18:51:01 -05:00
Henk
a87d5d6f23 Remove HF's llama workaround 2023-05-03 20:18:40 +02:00
Llama
35d344b951 Remove torch dependency and more generic dim0 workaround 
Remove torch dependency from hf.py
Make workaround for dimension zero values of token_ids
more generic to handle every token, not just newlines.
 2023-05-03 09:48:16 -07:00
Llama
3768848548 Fix tokenization and whitespace issues with llama-derived models 
Work around the 'soft' prefix space behavior of sentencepiece.
Override encode to restore the deleted HF support for decode_with_prefix_space.
Override decode to skip the soft space and return true decoded tokens.
Allow submitting chat messages with embedded newlines.
Split sentences between punctuation and whitespace, rather than after whitespace.
Also include trailing quotes and brackets after sentence stoppers.
This avoids splitting ." and .) into two tokens, for instance.
Insert whitespace at the beginning of the author's note, since sentences are
split with leading whitespace.
Remove spurious newlines at the end of chat responses.
 2023-05-03 01:27:11 -07:00
somebody
a0f4ab5c6a Move bad token grabber until after newlinemode has been deduced 2023-05-02 20:23:36 -05:00
somebody
efe268df60 Move overrides to better places 2023-05-02 20:18:33 -05:00
somebody
f6b5548131 Support safetensors in get_sharded_checkpoint_num_tensors 2023-05-01 19:15:27 -05:00
somebody
97e84928ba Download all shards correctly on aria2 and raise on bad load key 2023-05-01 18:53:36 -05:00
somebody
933dbd634a HFInferenceModel: Make badwordsids not unique to torch 2023-05-01 17:13:33 -05:00
somebody
ce3d465972 Remove some debug 2023-05-01 17:03:34 -05:00
onesome
467f2f25eb More loading fixes 2023-04-26 16:58:33 -05:00
onesome
d4f7b60dc9 Fix for multiple paths 2023-04-26 16:49:12 -05:00