diff options
| author | Carolinabanana <140120812+Carolinabanana@users.noreply.github.com> | 2024-04-09 09:16:13 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-04-09 11:16:13 +0300 |
| commit | 5dc9dd7152dedc6046b646855585bd070c91e8c8 (patch) | |
| tree | d2bae3652d91cdd9327e28fa85d167a67e050c53 /ggml-cuda/convert.cuh | |
| parent | e11a8999b5690f810c2c99c14347f0834e68c524 (diff) | |
llama : add Command R Plus support (#6491)
* Add Command R Plus GGUF
* Add Command R Plus GGUF
* Loading works up to LayerNorm2D
* Export new tensors in 1D so they are not quantized.
* Fix embedding layer based on Noeda's example
* Whitespace
* Add line
* Fix unexpected tokens on MPS. Re-add F16 fix. ((Noeda)
* dranger003: Fix block index overflow in CUDA dequantizing.
* Reverted blocked multiplication code as it still has issues and could affect other Llama arches
* export norms as f32
* fix overflow issues during quant and other cleanup
* Type convention
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* dranger003: Fix more int overflow during quant.
---------
Co-authored-by: S <seast@Ss-Mac-Studio.local>
Co-authored-by: S <s@example.com>
Co-authored-by: slaren <slarengh@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Diffstat (limited to 'ggml-cuda/convert.cuh')
| -rw-r--r-- | ggml-cuda/convert.cuh | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/ggml-cuda/convert.cuh b/ggml-cuda/convert.cuh index db34c0be..5394be9f 100644 --- a/ggml-cuda/convert.cuh +++ b/ggml-cuda/convert.cuh @@ -3,7 +3,7 @@ #define CUDA_DEQUANTIZE_BLOCK_SIZE 256 template<typename T> -using to_t_cuda_t = void (*)(const void * __restrict__ x, T * __restrict__ y, int k, cudaStream_t stream); +using to_t_cuda_t = void (*)(const void * __restrict__ x, T * __restrict__ y, int64_t k, cudaStream_t stream); typedef to_t_cuda_t<float> to_fp32_cuda_t; typedef to_t_cuda_t<half> to_fp16_cuda_t; |