From 8841ce3f439de6e770f70319b7e08b6613197ea7 Mon Sep 17 00:00:00 2001 From: Georgi Gerganov Date: Sun, 27 Oct 2024 20:59:58 +0200 Subject: [PATCH] llama : switch KQ multiplication to F32 precision by default (#10015) ggml-ci --- src/llama.cpp | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) diff --git a/src/llama.cpp b/src/llama.cpp index 50eebc2c2..53979e83f 100644 --- a/src/llama.cpp +++ b/src/llama.cpp @@ -9618,20 +9618,16 @@ static struct ggml_tensor * llm_build_kqv( cur = ggml_flash_attn_ext(ctx, q, k, v, kq_mask, kq_scale, hparams.f_max_alibi_bias, hparams.attn_soft_cap ? hparams.f_attn_logit_softcapping : 0.0f); - if (model.arch == LLM_ARCH_PHI2 || model.arch == LLM_ARCH_PHI3 || model.arch == LLM_ARCH_GPTNEOX || model.arch == LLM_ARCH_GEMMA2) { - ggml_flash_attn_ext_set_prec(cur, GGML_PREC_F32); - } + ggml_flash_attn_ext_set_prec(cur, GGML_PREC_F32); cur = ggml_reshape_2d(ctx, cur, n_embd_head_v*n_head, n_tokens); } else { struct ggml_tensor * kq = ggml_mul_mat(ctx, k, q); cb(kq, "kq", il); - if (model.arch == LLM_ARCH_PHI2 || model.arch == LLM_ARCH_PHI3 || model.arch == LLM_ARCH_GPTNEOX || model.arch == LLM_ARCH_QWEN2 || model.arch == LLM_ARCH_NEMOTRON || model.arch == LLM_ARCH_CHATGLM) { - // for this arch, we need to perform the KQ multiplication with F32 precision, otherwise we get NaNs - // ref: https://github.com/ggerganov/llama.cpp/pull/4490#issuecomment-1859055847 - ggml_mul_mat_set_prec(kq, GGML_PREC_F32); - } + // note: this op tends to require high floating point range + // while for some models F16 is enough, for others it is not, so we default to F32 here + ggml_mul_mat_set_prec(kq, GGML_PREC_F32); if (model.arch == LLM_ARCH_GROK) { // need to do the following: @@ -9640,9 +9636,6 @@ static struct ggml_tensor * llm_build_kqv( // kq = 30 * tanh(kq / 30) // before the softmax below - //try from phi2 - //ggml_mul_mat_set_prec(kq, GGML_PREC_F32); - kq = ggml_tanh(ctx, ggml_scale(ctx, kq, 0.08838834764831845f/30.0f)); kq = ggml_scale(ctx, kq, 30); }