-
### Issue summary
The estimated gradients by `GradientChecker` are not correct. I implemented my own Caffe layer and the gradients are the same as [PyTorch](https://github.com/pytorch/pytorch) and …
hzxie updated
5 years ago
-
I found a C64 core synchronization problem in Space is Broken - great demo by FAIRLIGHT.
In some scenes, artifacts appear that are not visible on real hardware or in the VICE emulator.
[https://csd…
F-RX updated
8 months ago
-
Single Op LLKs
- [x] #2258
- [x] #2259
- [x] #2283
- [x] #2292
Fused Op LLKs
- [x] #2260
- [x] #2261
- [x] #2557
Op Failure
- [x] rsqrt - #2315
- [x] power op #2314
- [ ] sign-bit #231…
-
**Your question**
Getting OOM errors training Megatron LM on Phi3 architecture. Should that happen?
```
GPU = A100 80GB (1 node)
minibatchsize = 1
sequence length = 4096
hidden size = 3072
hid…
-
**Describe the bug**
I am testing [demos for WH](https://github.com/tenstorrent/tt-metal/tree/main#wormhole-wh-models) on N150. And encountering errors.
**To Reproduce**
Steps to reproduce the b…
-
We currently only support 2D inputs for matmul. Extend implementation to support additional cases:
1. Both A and B are 1D (`[M,] x [M,]`)
2. A is 1D and B is 2D (`[M,] x [M, N]`)
4. A is 2D and B…
-
Getting an error when running mistrual7b.
```
pytest --disable-warnings -q -s --input-method=cli --cli-input="YOUR PROMPT GOES HERE!" models/demos/mistral7b/demo/demo.py
2024-05-03 05:17:51.051 | …
-
The `device` argument is just ignored
https://github.com/ndif-team/nnsight/blob/320a9b702bf264b87e354494c18ff8cd2646f518/src/nnsight/models/UnifiedTransformer.py#L37
-
This issue is to discuss Liisa question on adding DFO's functional areas to the MRF form to help in reporting on the main area addressed by the manuscript.
- [x] Should we do it?
- [x] Get list of…
-
### 🚀 The feature
Move checks for zeros and tensor creation to __init__ of torchvision transform normalize
https://pytorch.org/vision/main/_modules/torchvision/transforms/transforms.html#Normalize…