Comments (2)
@gokulavasan @joecummings I'm being a bit pedantic but I htink this has more to do with reproducibility rather than determinism (the nuance is important)
from torchtune.
@gokulavasan @joecummings I'm being a bit pedantic but I htink this has more to do with reproducibility rather than determinism (the nuance is important)
Good point. This is for reproducibility.
from torchtune.
Related Issues (20)
- safe_torch_load failed when resume from checkpoint HOT 6
- [ignore] testing issue
- Resize token embedding. HOT 1
- generate is correct but generate from quantization get error: HOT 5
- [feature request] Saving / Loading packed dataset HOT 6
- Changing Linear layer to ParamList cause significant increase in memory usage in Llama3-8B LoRA fine-tuning. HOT 3
- lm-eval==0.4.3 breaks euleuther_eval.py script HOT 2
- Shape error when using torchtune.modules.RotaryPositionalEmbeddings HOT 3
- ALLGATHER_BASE timeout error HOT 7
- Checkpoint save in FSDP2 recipe hangs on 4 devices but not on 2 devices HOT 5
- Mask eos token for packed dataset HOT 5
- load_dataset fails on distributed recipes for datasets with remote code HOT 2
- Model merging scripts? HOT 1
- [RFC] Adding RoPE scaling methods to support long context modeling HOT 10
- expandable_segments with PYTORCH_CUDA_ALLOC_CONF reduces VRAM HOT 2
- torch.compile support HOT 2
- Error loading state dict using llama3_lora for inference HOT 8
- Error resuming lora training using config if checkpoint_dir is different from output_dir HOT 2
- `wikitext_dataset` fails in model forward with "Expected Long, Int; but got torch.cuda.FloatTensor" HOT 1
- Modify ChatML chat format to support generation
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from torchtune.