-
## 🚀 Feature
It would be useful if we would model constants and use that to optimize away `getitem`.
### Motivation
The ThunderFX path sometimes ends up giving us some rather silly graphs:
…
-
**Describe the bug**
The function `ttnn.softmax` exhibits numerical instability when handling input tensors with large values. Specifically, when provided with such inputs, the function produces in…
-
```python
import alpa
import numpy as np
import jax.numpy as jnp
@alpa.parallelize
def tensor_compute(tensor):
t_transposed = jnp.transpose(tensor)
dot_matrix = jnp.dot(tensor, …
-
https://dl.acm.org/doi/10.1145/3582016.3582018
[ASPLOS 2023: Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3]…
-
Hello,
I measured the time of your BitLinear and BitLinearBitBLAS against nn.Linear, and it seems that the time for smaller input_features and out_features is slower than nn.Linear. Is there a solu…
-
I am running the Ubuntu 22 DLAMI, and trying to profile a simple kernel that performs exponentiation:
```
"""
Example kernel used to demmonstrate Neuron Profile.
"""
import torch
import neuronxc…
-
---
### Bug Report: In-Place Operation Causes Gradient Error in `conv1d_step` Function
**Issue Description:**
While training the model, I encountered a runtime error related to gradient compu…
-
### What happened?
For the give IR
```mlir
#map = affine_map ()>
#map1 = affine_map (d0, d1, d2, d3)>
#map2 = affine_map ()>
#map3 = affine_map (s0 floordiv 12)>
module {
ml_program.global…
-
Hello,
I am encountering an issue related to my understanding of tensor parallelism in the PIM (Processing In Memory) model.
Specifically, I noticed a discrepancy in the Key-Value (KV) cache all…
-
thanks for ur project.
when i run GBLM-Pruner/gradient_computation.py, I meet a error whose details are below.
Traceback (most recent call last):
File "GBLM-Pruner/gradient_computation.py", lin…