From 85c5a1a99569ccc00c280835fe3a69b4af02c43b Mon Sep 17 00:00:00 2001 From: Kawrakow Date: Sun, 15 Dec 2024 09:54:21 +0100 Subject: BF16_R16 - 16 interleaved bf16 rows (#142) * Not working bf16_r4 * Adding bf16_r8 Small performance gain compared to bf16 - 258 t/s vs 234 t/s. I guess, this is still sub-obtimal. * bf16_rx: Very slightly faster by interleaving 16 rows 258 t/s -> 263 t/s * Rename bf16_r4 to bf16_r16 We are interleaving 16 rows now. * Cleanup unused stuff --------- Co-authored-by: Iwan Kawrakow --- examples/quantize/quantize.cpp | 1 + 1 file changed, 1 insertion(+) (limited to 'examples') diff --git a/examples/quantize/quantize.cpp b/examples/quantize/quantize.cpp index 0f906b83..5e5dd7c0 100644 --- a/examples/quantize/quantize.cpp +++ b/examples/quantize/quantize.cpp @@ -77,6 +77,7 @@ static const std::vector QUANT_OPTIONS = { { "Q4_0_8_8", LLAMA_FTYPE_MOSTLY_Q4_0_8_8, " 4.34G, +0.4685 ppl @ Llama-3-8B", }, { "F16", LLAMA_FTYPE_MOSTLY_F16, "14.00G, -0.0020 ppl @ Mistral-7B", }, { "BF16", LLAMA_FTYPE_MOSTLY_BF16, "14.00G, -0.0050 ppl @ Mistral-7B", }, + { "BF16_R16", LLAMA_FTYPE_MOSTLY_BF16_R16, "14.00G, -0.0050 ppl @ Mistral-7B", }, { "F32", LLAMA_FTYPE_ALL_F32, "26.00G @ 7B", }, // Note: Ensure COPY comes after F32 to avoid ftype 0 from matching. { "COPY", LLAMA_FTYPE_ALL_F32, "only copy tensors, no quantizing", }, -- cgit v1.2.3