Skip to content

Issues: pytorch/torchtune

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

91 Open 273 Closed
91 Open 273 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

expandable_segments with PYTORCH_CUDA_ALLOC_CONF reduces VRAM
#1185 opened Jul 16, 2024 by winglian
1
[RFC] Adding RoPE scaling methods to support long context modeling rfc
#1183 opened Jul 16, 2024 by joecummings
3
Model merging scripts?
#1179 opened Jul 15, 2024 by suraj-srinivas
@ebsmothers
1
load_dataset fails on distributed recipes for datasets with remote code bug Something isn't working
#1178 opened Jul 15, 2024 by pbontrager
@RdoubleA
1
Mask eos token for packed dataset
#1177 opened Jul 15, 2024 by iankur
3
ALLGATHER_BASE timeout error
#1165 opened Jul 11, 2024 by aknvictor
6
[feature request] Saving / Loading packed dataset enhancement New feature or request help wanted Extra attention is needed
#1149 opened Jul 8, 2024 by ScottHoang
3
generate is correct but generate from quantization get error: help wanted Extra attention is needed question Further information is requested
#1148 opened Jul 8, 2024 by artisanclouddev
5
Resize token embedding. help wanted Extra attention is needed question Further information is requested
#1145 opened Jul 7, 2024 by hungphongtrn
@ebsmothers
1
Shape error when using torchtune.modules.RotaryPositionalEmbeddings question Further information is requested
#1157 opened Jul 6, 2024 by Leo-Lifeblood
@ebsmothers
3
safe_torch_load failed when resume from checkpoint bug Something isn't working question Further information is requested
#1142 opened Jul 3, 2024 by ScottHoang
@ebsmothers
6
text_completion_dataset removed? question Further information is requested
#1140 opened Jul 3, 2024 by wiiiktor
@joecummings
5
Optimize attention mask creation for sample packing.
#1130 opened Jun 27, 2024 by joecummings
Quantization for Llama-70b raises CUDA OOM question Further information is requested
#1128 opened Jun 27, 2024 by lulmer
@ebsmothers
2
How to finetune a model for sequence classification tasks using the dataset like stanfordnlp/imdb from huggingface?
#1124 opened Jun 26, 2024 by JonasQN
1
Improve dataset documentation
#1123 opened Jun 26, 2024 by RdoubleA
2
Support for Phi-3-mini-128k-instruct and larger context length models
#1120 opened Jun 25, 2024 by dcsuka
@joecummings
Add in special tokens for Gemma and Mistral in tokenizer
#1118 opened Jun 25, 2024 by joecummings
CUDA illegal memory access exception
#1112 opened Jun 24, 2024 by l3utterfly
5
Missing non-LoRA key tok_embeddings.weight from base model dict
#1110 opened Jun 22, 2024 by vasicvuk
2
Save intermediate checkpoints during training question Further information is requested
#1107 opened Jun 21, 2024 by l3utterfly
@joecummings
4
want dora and nef-tune supports! enhancement New feature or request
#1100 opened Jun 20, 2024 by jeffchy
@calvinpelletier
3
How to provide our own daaset.json or csv file to the llama2:7B finetune using tune run
#1099 opened Jun 19, 2024 by himanshushukla12
13
Support NF4 quantization of linear layers without LoRA applied
#1093 opened Jun 14, 2024 by ebsmothers
6
High memory usage on Llama3-70B full finetune during checkpoint save
#1092 opened Jun 14, 2024 by ebsmothers
@ebsmothers
1
ProTip! Follow long discussions with comments:>50.