Commit Graph

190 Commits

Author SHA1 Message Date
bfe03dbdc2 WIP cdata and comment tokens 2025-04-13 22:02:29 +02:00
773134e65b Improved tests for tokens 2025-04-13 19:05:10 +02:00
ac2c1ae08b Added lib to clean-python in Makefile 2025-04-13 18:19:32 +02:00
5bcfe89cb9 Disabled community cloud 2025-04-13 18:15:19 +02:00
be48130a8f Use CharTrie and fix multi-byte char issue 2025-04-13 16:05:27 +02:00
f4de9e7571 Fixed token cache for whitespace bug 2025-04-13 10:03:14 +02:00
ff82173f2a Removed auto enable for QwQ 2025-04-13 09:56:01 +02:00
c2bf422363 Fixed whitespace processing 2025-04-12 19:59:25 +02:00
9564879edc Fixed attribute parsing 2025-04-12 18:02:55 +02:00
bd3adf78e6 Fix attributes in self-closing tags 2025-04-11 19:26:05 +02:00
7e254b7f7e Fixed qwq token limit 2025-04-11 18:03:17 +02:00
Niels Geens
e9f5bcc519 remove debug print 2025-04-10 22:10:29 +02:00
Niels Geens
ba9d69b414 Added char trie 2025-04-10 21:46:24 +02:00
Niels Geens
10df42c3a6 Added small optimization to logits processor core 2025-04-10 19:11:59 +02:00
Niels Geens
d81ae09f67 Fix nested elements 2025-04-10 18:37:00 +02:00
Niels Geens
2f0fd21b0a Fix missing attribut 2025-04-10 18:14:25 +02:00
Niels Geens
cfb65ee710 Handle attributes and self-closing tags 2025-04-10 14:35:14 +02:00
Niels Geens
5b8f04be81 fix copy bug 2025-04-09 16:43:27 +02:00
e1da61fd71 Fix import for xml_schema_validator 2025-04-09 10:56:20 +00:00
Niels Geens
8fb30c8ed0 Optimized hot-loop for logits processor 2025-04-09 11:59:26 +02:00
654c32c7ac wip logits processor in ruuuuuuust 2025-04-08 06:34:03 +00:00
3826c45240 Made xml schema validator send 2025-04-07 19:27:03 +02:00
Niels Geens
2d4a57a706 Changed standard output of collect.sh to collect.txt 2025-04-07 16:52:12 +02:00
2e08be4b75 Validator works for start 2025-04-07 14:31:56 +00:00
Niels Geens
50539f6d03 Integrate logits processor with qwq (untested) 2025-04-07 13:35:20 +02:00
Niels Geens
023fc75aac Merge branch 'master' of https://git.nielsgeens.be/llm/SIA 2025-04-07 12:19:26 +02:00
Niels Geens
faf78b7948 Lock vllm version to avoid bug 2025-04-07 12:17:56 +02:00
Niels Geens
f4a17ad01e Lock vllm version to avoid bug 2025-04-07 11:57:15 +02:00
Niels Geens
cda067eb07 Removed jupyter venvs for all but train 2025-04-07 11:43:20 +02:00
Niels Geens
9c2d38e1a6 Fix install script, use correct venvs for jupyter 2025-04-07 11:36:41 +02:00
ead5c35a09 Added rust dependency to docker build 2025-04-06 22:41:34 +02:00
192ade72ee Fixed build on cloud 2025-04-06 14:46:42 +00:00
ea93adc4c8 Fix dockerfile for xml schema validator 2025-04-06 16:11:36 +02:00
55a920414d Begin XML Schema Validator 2025-04-05 23:15:39 +02:00
0e2cca2e4f Fixed QwQ tokenizer 2025-04-04 08:54:06 +00:00
156d3aa7d3 Bootstrapping QwQ works!! 2025-04-02 16:37:17 +00:00
Niels Geens
b4c08586ee Applied add_venvs_to_path.sh script to local runs 2025-04-02 18:05:48 +02:00
ed37792b62 Better path fix 2025-04-02 15:58:30 +00:00
ea22e1527e Fix bug where runpod environment file overwrites PATH 2025-04-02 14:47:33 +00:00
Niels Geens
e573b9ebe7 Removed flash-attn as dependency because it's hard to install properly 2025-04-02 14:42:35 +02:00
78c71ab8b4 Improved install experience 2025-04-01 15:26:50 +00:00
Niels Geens
c00256f723 Clean up install script 2025-04-01 16:30:42 +02:00
130afbdd92 Updates after testing 2025-04-01 11:49:21 +00:00
Niels Geens
6f3d414d17 Converted QwQ notebooks to .py files 2025-04-01 10:42:54 +02:00
642c5181bb Fixed inference notebook 2025-03-30 08:52:17 +00:00
290c718187 Added qwq inference notebook 2025-03-30 09:26:18 +02:00
30eb5c419d Updated QwQ train notebook 2025-03-29 11:03:53 +00:00
585f655ad3 Better error handling when sourcing bootstrap script 2025-03-29 10:38:25 +01:00
56242496bb Add support for jupyter notebooks for training 2025-03-28 16:10:43 +00:00
468ef2ae1f configure git username and email in bootstrap script 2025-03-28 15:09:37 +00:00