Comments (1)
One Update
I tried locally compiling with string(APPEND CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -fmad=false")
And the error still repros
Also while using: NanInfdetect
I was able to determine that this training loops is encouraging very large values for q, k , v
4.9956e+08, -4.9956e+08, -4.9956e+08, ..., -4.9956e+08,
-4.9956e+08, -4.9956e+08],
[ 2.7200e+08, 2.7200e+08, 2.7200e+08, ..., 2.7200e+08,
2.7200e+08, 2.7200e+08],
If you set norm_first = True
I am unable to reproduce the error, the fused kernels are known to be more sensitive to large values, I have seen this directly for FAv2 because of its use of FMA.. I don think thats the case here from the first update though.
from pytorch.
Related Issues (20)
- [aot autograd][inline inbuilt nn modules] AOT Autograd changes _version different from eager HOT 6
- torch.utils.flop_counter.FlopCounterMode doesn't work well with compiled model
- Wrong result for Inplace tensor update on transpose for some devices with torch 2.3.0 HOT 2
- Documentation for DDP-related environment variables HOT 1
- [Bug] The cuDNN version is too old! HOT 9
- Adam is storing step as cpu tensor
- nn.Transformer gives different output in torch.no_grad() context HOT 3
- Memory leak when assigning function to instance variable HOT 6
- DISABLED test_attention_vs_linear (__main__.MemoryBudgetTest) HOT 2
- `torch.compile` with `reduce-overhead` and DDP causes crash in inference after training HOT 1
- Why does the size of state_dict decrease a lot when I move the parameters to the CPU? What information was lost? HOT 6
- Rebase your PRs: Unstable CUDA signal in CI caused by cudnn 9 update HOT 1
- torch.jit.script causes Segmentation fault 139 HOT 3
- AOTAutogradCache implementation
- Cannot convert -oo to int HOT 4
- 'FakeRootModule' object has no attribute 'self___aot_engines_0_short_term_memories_list_0_0_0' HOT 4
- Use of standard library math functions vs global namespace
- can not create custom gradient when fp8 type is used HOT 2
- JIT script can cause unneeded `requires_grad` inside `no_grad` blocks
- torchinductor error in torchao tests HOT 5
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 pytorch.