diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2024-12-17 14:16:34 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-12-17 14:16:34 +0100 |
| commit | 514ae086200a8cfd78af6a71b6c6ee14931ddc0e (patch) | |
| tree | 0fa47186d7c82afbf078d530f5436c7eb1ae4d79 /common/common.h | |
| parent | 4ade4c568c331acad22537f7b9519c740c7a06d0 (diff) | |
Be able to repack tensors at run time (#147)
* Be able to repack tensors at run time
* Repack: also add bf16 as repackable type
* Repack: make sure number of rows is a multiple of the packing
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'common/common.h')
| -rw-r--r-- | common/common.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/common/common.h b/common/common.h index 486017ef..73d7d650 100644 --- a/common/common.h +++ b/common/common.h @@ -187,6 +187,7 @@ struct gpt_params { bool no_kv_offload = false; // disable KV offloading bool warmup = true; // warmup run bool check_tensors = false; // validate tensor data + bool repack_tensors = false; // repack tensors if interleaved variant is available std::string cache_type_k = "f16"; // KV cache data type for the K std::string cache_type_v = "f16"; // KV cache data type for the V |