Skip to content

Issues: pytorch/torchtune

v0.6.0 tracker
#2232 opened Jan 6, 2025 by joecummings
Open
Testing tracker
#1890 opened Oct 23, 2024 by felipemello1
Open
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

266 Open 614 Closed
266 Open 614 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

Incorrect Default Config File Paths for Llama 3.1 8B and Qwen 2.5 7B Models
#2352 opened Feb 6, 2025 by MaxHastings
1
"ft-" prefix for finetuned checkpoints
#2345 opened Feb 5, 2025 by EugenHotaj
3
Discussion: Update dataloader to skip rows that dont require training best practice Things we should be doing but aren't discussion Start a discussion triage review This issue should be discussed in weekly review
#2344 opened Feb 5, 2025 by felipemello1
3
CEWithChunkedOutputLoss does not check division by zero discussion Start a discussion triaged This issue has been assigned an owner and appropriate label
#2341 opened Feb 4, 2025 by pocca2048
@joecummings
6
Feature request: GRPO support
#2340 opened Feb 4, 2025 by tikikun
4
Seed: null isn't random bug Something isn't working triaged This issue has been assigned an owner and appropriate label
#2338 opened Feb 3, 2025 by bogdansalyp
1
Seed is not applied for DPO recipes bug Something isn't working triaged This issue has been assigned an owner and appropriate label
#2335 opened Feb 3, 2025 by bogdansalyp
3
Apply gradient accumulation fix to DPO/PPO recipes
#2334 opened Feb 3, 2025 by bogdansalyp
1
Distributed DPO loss normalization by amount of tokens
#2333 opened Feb 3, 2025 by bogdansalyp
2
Wandb charts show time (minutes), but I want seconds.
#2329 opened Feb 2, 2025 by kailashg26
1
FIRE Relative Positional Encodings
#2325 opened Jan 31, 2025 by kaddu341
2
Reading TorchProfiler after run
#2323 opened Jan 31, 2025 by fabiogeraci
6
Classifiers (reward models) in torchtune
#2320 opened Jan 30, 2025 by EugenHotaj
4
How to run torchtune on AMD Instinct MI300X
#2319 opened Jan 30, 2025 by kailashg26
2
Call get_world_size_and_rank ONCE
#2316 opened Jan 29, 2025 by joecummings
Rename and document cleanup_before_training
#2315 opened Jan 29, 2025 by joecummings
Investigate the optimal scenario in which to use torch_set_num_thread()
#2312 opened Jan 29, 2025 by joecummings
Unable to reproduce QAT results from Blog
#2310 opened Jan 29, 2025 by AbhinavDutta
9
Differing component implementation logic across recipes best practice Things we should be doing but aren't better engineering Tasks which help improve eng productivity e.g. building tools, cleaning up code, writing docs bug Something isn't working triaged This issue has been assigned an owner and appropriate label
#2307 opened Jan 29, 2025 by EugenHotaj
3
Support for Janus-Pro series of model
#2306 opened Jan 28, 2025 by Ankur-singh
2
Move from PIL to torchvision.io.decode_image best practice Things we should be doing but aren't community help wanted We would love the community's help completing this issue
#2303 opened Jan 28, 2025 by ebsmothers
@Ankur-singh
3
Missing <|begin_of_text|> Token in Llama3Tokenizer
#2300 opened Jan 27, 2025 by seungjun-green
3
Tokens per second calculation best practice Things we should be doing but aren't triage review This issue should be discussed in weekly review
#2296 opened Jan 24, 2025 by EugenHotaj
@EugenHotaj
8
Roadmap for distributed recipes using NPU as a backend
#2288 opened Jan 22, 2025 by Nicorgi
8 tasks
deepseek r1 support? enhancement New feature or request triage review This issue should be discussed in weekly review
#2287 opened Jan 21, 2025 by johnnynunez
10
ProTip! Type g i on any issue or pull request to go back to the issue listing page.