Skip to content

Pull requests: axolotl-ai-cloud/axolotl

New pull request New
67 Open 1,108 Closed
67 Open 1,108 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

shampoo optim support
#1919 opened Sep 18, 2024 by winglian Loading…
Grokfast support
#1917 opened Sep 17, 2024 by winglian Loading…
Add Support for revision Dataset Parameter to specify reading from Huggingface Dataset Revision
#1912 opened Sep 12, 2024 by thomascleberg Loading…
3
support seperate lr for embeddings, similar to loraplus
#1910 opened Sep 12, 2024 by winglian Loading…
Refactor func load_model to class ModelLoader
#1909 opened Sep 12, 2024 by MengqingCao Loading…
1 task
wip add new proposed message structure wip
#1904 opened Sep 7, 2024 by winglian Loading…
Add Llama 3.1 Instruct Chat template, Ensure Correctness of 3, Small Refactor of Chat Template registration for shareGPT
#1903 opened Sep 7, 2024 by alpayariyak Loading…
2
add ds zero3 to multigpu biweekly tests
#1900 opened Sep 5, 2024 by winglian Loading…
Reward model
#1879 opened Aug 28, 2024 by winglian Loading…
phi moe support for multipack
#1870 opened Aug 26, 2024 by winglian Loading…
drop valueerror as this was from when 4bit required gptq
#1857 opened Aug 22, 2024 by winglian Loading…
1
semi-weekly 8bit lora zero3 check
#1852 opened Aug 22, 2024 by winglian Loading…
fix so inference can be run against quantized models without adapters
#1834 opened Aug 19, 2024 by winglian Loading…
examples: Fix config llama3
#1833 opened Aug 19, 2024 by JohanWork Loading…
1
[DO NOT MERGE] bump accelerate and transformers to main hold don't merge this yet
#1764 opened Jul 17, 2024 by winglian Loading…
Enable Ascend NPU support
#1758 opened Jul 16, 2024 by MengqingCao Loading…
6
add q-galore optimizer
#1752 opened Jul 14, 2024 by winglian Loading…
3
Implements SPPO Alignment Algoritm
#1735 opened Jul 11, 2024 by kaykyr Loading…
1 of 3 tasks
10
remove the bos token from dpo outputs
#1733 opened Jul 10, 2024 by winglian Loading…
1
1
Allow using tokenizer's default chat template or pass custom jinja chat template
#1732 opened Jul 9, 2024 by chiragjn Loading…
1
44
Update multi-node.qmd
#1688 opened Jun 7, 2024 by shahdivax Loading…
15
update to be deprecated evaluation_strategy and c4 dataset
#1682 opened Jun 3, 2024 by winglian Loading…
jagged lr restart scheudler
#1680 opened Jun 3, 2024 by winglian Loading…
add support for SPPO
#1585 opened May 2, 2024 by winglian Loading…
WIP test out new dockerfile with more nvidia tools
#1557 opened Apr 21, 2024 by winglian Loading…
ProTip! Adding no:label will show everything without a label.