Skip to content

Issues: Lightning-AI/litgpt

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

260 Open 589 Closed
260 Open 589 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

Mistral7B pretraining convergence is too slow question Further information is requested
#697 opened Nov 5, 2023 by LamOne1
5
Significantly different results with inference using a saved checkpoint v/s inferencing during fine-tuning
#686 opened Oct 30, 2023 by madhurapande19
2
[question] how to set the max_iters value question Further information is requested
#682 opened Oct 29, 2023 by nevermet
1
[question] how to finetune efficiently question Further information is requested
#680 opened Oct 28, 2023 by nevermet
4
Custom 4k context length supporting and converting model config to huggingface supportted config file question Further information is requested
#666 opened Oct 24, 2023 by SinclairCoder
2
Finetuning bug bug Something isn't working
#645 opened Oct 15, 2023 by Gideonah
1
Add NEFTune to improve fine-tuning? enhancement New feature or request fine-tuning
#642 opened Oct 13, 2023 by nkasmanoff
4
AttributeError: <class 'lit_gpt.utils.NotYetLoadedTensor'> does not have permute bug Something isn't working
#639 opened Oct 11, 2023 by azinman
Any finetuning is having No effect on small models question Further information is requested
#627 opened Oct 8, 2023 by VatsaDev
2
Sample packing for pretraining/fine-tuning enhancement New feature or request pre-training
#620 opened Oct 6, 2023 by alitirmizi23
16
Will longlora finetune supported in the future? model-weights
#616 opened Oct 5, 2023 by universewill
2
QA-LoRA: Quantization Aware Low-Rank Adaptation enhancement New feature or request fine-tuning
#595 opened Sep 28, 2023 by Andrei-Aksionov
3
How to use Activation Checkpointing and Parameter Offloading in a single GPU? bug Something isn't working
#594 opened Sep 28, 2023 by ifshine
Add tutorial for custom checkpoints model-weights
#591 opened Sep 27, 2023 by JOHW85
2
[Question] How to decrease my loss? question Further information is requested
#575 opened Sep 21, 2023 by ifshine
6
Adding RLHF support enhancement New feature or request fine-tuning
#557 opened Sep 16, 2023 by rasbt
4
Can I use Wizardcoder to finetune？ model-weights
#555 opened Sep 16, 2023 by hello-eternity
1
Can I use lora or adapter to fine-tune some non-instruction set data? question Further information is requested
#550 opened Sep 15, 2023 by hello-eternity
3
slow finetuning on TPUv3-8 bug Something isn't working
#519 opened Sep 10, 2023 by ethanhe42
TPU RuntimeError: Expected query, key, and value to have the same dtype, but got query.dtype: float key.dtype: c10::BFloat16 and value.dtype: c10::BFloat16 instead. bug Something isn't working
#518 opened Sep 9, 2023 by ethanhe42
2
LoRA with quantization: micro_batch_size effect on memory footprint bug Something isn't working quantization
#501 opened Sep 6, 2023 by Andrei-Aksionov
3 tasks
3
OOM with bf16-true, Quantization, for long context length. bug Something isn't working quantization
#477 opened Aug 29, 2023 by KOVVURISATYANARAYANAREDDY
18
QLoRA for Multi-GPU Settings enhancement New feature or request fine-tuning
#449 opened Aug 22, 2023 by rasbt
4
Look into adding Dolma pretraining dataset enhancement New feature or request
#439 opened Aug 18, 2023 by rasbt
Evaluation with triviaqa bug Something isn't working evaluation
#424 opened Aug 16, 2023 by sanyalsunny111
ProTip! Add no:assignee to see everything that’s not assigned.