-
在处理好了PG19数据之后,进行训练,一直发现有问题
[WARNING|logging.py:329] 2024-05-14 16:24:22,784 >> LlamaModel is using LlamaSdpaAttention, but `torch.nn.functional.scaled_dot_product_attention` does not support `output_…
-
https://zhuanlan.zhihu.com/p/338817680
-
**What is your question?**
when I use sm89
```
int run_attention(Options& options) {
using Attention = AttentionKernel<
cutlass::half_t, // scalar_t
cutlass::arch::Sm89, // ArchT…
-
### Your current environment
Collecting environment information...
/opt/conda/envs/py_3.9/lib/python3.9/site-packages/torch/cuda/__init__.py:611: UserWarning: Can't initialize NVML
warnings.warn(…
-
File "torch/_dynamo/variables/constant.py", line 176, in call_method
return ConstantVariable.create(op(self.value, add_target))
torch._dynamo.exc.InternalTorchDynamoError: 'bool' object is n…
-
Platforms: dynamo, linux, rocm, asan
cc @bdhirsh @drisspg @mikaylagawarecki @NmomoN @mengpenghui @fwenguang @cdzhan @1274085042 @PHLens @jeffdaily @sunway513 @pruthvistony @ROCmSupport @dllehr-…
-
Not sure if bug or feature will draw attention to it anyway :)
![image](https://github.com/user-attachments/assets/87f8410f-fd31-4dd2-819d-ccc662e052f6)
-
suggested by Morgan:
Is it possible that the image getting the most attention on the day might be able to pop into some part of the front page?
-
Please take the time to add users as admin on the Discord server there are many many scam links that need to be cleaned out
Thank you
-
Your improved attention module has deeply inspired my work, could you please share your EGSA code with me so that I can have a deeper understanding of your work?