Alex

MixinPreProcessorStandard incorrectly checks STATIC flag on @Unique fields, a crash happens

1

**Steps to reproduce**: 1. Create a Mixin like this (1.18.2, Mojang names, Fabric): ``` import net.minecraft.server.packs.repository.Pack; import org.spongepowered.asm.mixin.Mixin; import org.spongepowered.asm.mixin.Unique; @Mixin(Pack.class) public final class TestMixin { // Matches name of...

As I understand, all models in RWKV family were trained on Pile dataset. I'm concerned with possible lack of preprocessing of the dataset. [Pile paper](https://arxiv.org/pdf/2101.00027.pdf) states `To avoid leakage of...

A way to automatically calculate mem_size for creating ggml context

4

As far as I understand, before creating ggml context, we need to know exactly how much memory is needed, and pass this value into `ggml_init` through `ggml_init_params.mem_size`. On the one...

This issue is a request for porting [k-quants](https://github.com/ggerganov/llama.cpp/blob/master/k_quants.c) from [llama.cpp](https://github.com/ggerganov/llama.cpp) into [ggml](https://github.com/ggerganov/ggml). See [the original PR for k-quants](https://github.com/ggerganov/llama.cpp/pull/1684) for details on why these new formats are useful. [Similar issue for...

Implement quantization on-the-fly

2

This feature allows to quantize FP32/FP16 models on-the-fly to any other quantized format, without the need to explicitly run `quantize.py` and keep quantized models on disk. Intended use-case is having...