Skip to content

Issues: unslothai/unsloth

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

500 Open 494 Closed
500 Open 494 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
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Resizing tokenizer leads to missing end token and garbage response?
#1273 opened Nov 10, 2024 by Mark-DelGrande
Jupyter notebook: No module named 'unsloth'
#1272 opened Nov 10, 2024 by iwouldratherbeatthebeach
feat: add option for using ADOPT optimizer based on Taniguchi, Shohei, et al.
#1270 opened Nov 9, 2024 by Selich
cannot load some models via vllm
#1268 opened Nov 8, 2024 by yananchen1989
4
save_pretrained_merged ruins my model
#1267 opened Nov 8, 2024 by Romiroz
Couldn't build proto file into descriptor pool! Invalid proto descriptor for file "sentencepiece_model.proto": sentencepiece_model.proto: A file with this name is already in the pool.
#1266 opened Nov 8, 2024 by CurtiusSimplus
5
LoRA on Qwen 2.5 does not patch qkv matrices
#1265 opened Nov 8, 2024 by MinghaoYan
TypeError: SFTTrainer.__init__() got an unexpected keyword argument 'dataset_text_field'
#1264 opened Nov 8, 2024 by officialsahyaboutorabi
3
[ERROR BUG] 'triton.language' has no attribute 'cast'. Did you mean: 'cat'?
#1263 opened Nov 7, 2024 by arianyambao
5
train_on_responses_only doesn't work for Mistral models
#1262 opened Nov 7, 2024 by XiaomoWu
2
CAN'T LOAD: AttributeError: 'LlamaForCausalLM' object has no attribute 'update'
#1261 opened Nov 7, 2024 by yukiarimo
2
ValueError: Unsloth: Untrained tokens of [[128004]] found
#1258 opened Nov 7, 2024 by Hyfred
import unsloth causes error: pip install unsloth-zoo
#1252 opened Nov 6, 2024 by DaddyCodesAlot
3
issue when using default settings for training
#1251 opened Nov 6, 2024 by Ammar-Alnagar
3
[FIXED] AssertionError('initial value for logits error fixed - pending confirmation Fixed, waiting for confirmation from poster URGENT BUG Urgent bug
#1248 opened Nov 6, 2024 by daegonYu
9
RuntimeError: CUDA error during inference from saved lora weights
#1244 opened Nov 5, 2024 by danisharoonds
1
Dataset creation to use with unsloth fine tuning
#1243 opened Nov 5, 2024 by gaussiangit
1
Unsloth error unable to push to hub
#1242 opened Nov 5, 2024 by hung-ngm
1
how to only do lora on the lm_head? feature request Feature request pending on roadmap
#1241 opened Nov 5, 2024 by brando90
3
why is unsloth thinking I'm doing multi gpu optimization when I'm not?
#1240 opened Nov 5, 2024 by brando90
3
Fine tuned Llama3.1 does not support tools
#1239 opened Nov 4, 2024 by darkroasted
1
erorr
#1238 opened Nov 4, 2024 by werruww
4
RuntimeError: ptxas failed with error code 4294967295:
#1237 opened Nov 4, 2024 by heiheiheibj
2
Overlap matrix multiplication (needs tensor core) and other things like activation (needs cuda core and memory bandwidth) to speed up
#1233 opened Nov 3, 2024 by fzyzcjy
2
Why is memory bandwidth only half used? Is it possible we speed up by utilizing this?
#1230 opened Nov 1, 2024 by fzyzcjy
2
ProTip! Exclude everything labeled bug with -label:bug.