Andrei-Aksionov
Andrei-Aksionov
## 🚀 Feature Hi there 👋 From the main readme file I noticed that `Thunder` except custom kernels, but only the ones that are written in Trition. Is there a...
[Daniel Han](https://twitter.com/danielhanchen) in his [blogpost](https://unsloth.ai/blog/gemma-bugs) shared his discoveries of what is ~also~ wrong with the Gemma implementation. Some of them are only appliable to Keras, some of them affects PyTorch...
Hi there 👋 This PR changes placement of `query`, `key` and `value` weights in the combined QKV matrix. Intuitively one might assume that weights in the QKV matrix are combined...
Hi there 👋 I came across a new quantization technique called Half-Quadratic Quantization. Blog: https://mobiusml.github.io/hqq_blog/ GitHub: https://github.com/mobiusml/hqq/tree/master This approach performs quantization to [1|2|3|4|8]bit precision, but doesn't require calibration process (like...
Hi there 👋 It's a bit late response to #583 The task itself turned out to be quite large, so in order to speed up the process (and simplify life...
Hi there 👋 This PR adds functionality to dequantize the model that was initially quantized, in case anyone wants to use these weights elsewhere. The logic is to dequantize to...
*While experimenting with GitHub actions I deleted my fork (I know, I know) and thus all opened PRs were automatically closed. This PR mirrors #630.* Hi there 👋 This PR...
Hi there 👋 One might ask why do we need to specify encoding explicitly, when it should be by default `utf-8`. Well, not quite. > The default encoding is platform...
Hi there, Just want to share implementation of this repo but in python: https://github.com/Andrei-Aksionov/TheLastAlgorithmsCourseYouWillNeed Basically the same drill: implement algorithm, run tests. Stuck and don't know how to implement -...
Hello there 👋 Thanks for repo! Decided to add my little contribution. With the current main the logging output during quantization is such: ```text INFO - Start quantizing layer 1/6...