From 7e7cd53ca6f52ffffe35d44439f3db1c0ac4dfa8 Mon Sep 17 00:00:00 2001 From: Pierrick HYMBERT Date: Sat, 6 Apr 2024 23:55:37 +0200 Subject: [PATCH] llama: dbrx: remove unnecessary optional tensor on FFN_GATE_EXPS --- llama.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/llama.cpp b/llama.cpp index 7347ae655..b8ca64b65 100644 --- a/llama.cpp +++ b/llama.cpp @@ -4692,7 +4692,7 @@ static bool llm_load_tensors( layer.wqkv = ml.create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd}); layer.ffn_gate_inp = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_FFN_GATE_INP, "weight", i), {n_embd, n_expert}); - layer.ffn_gate_exps = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_GATE_EXPS, "weight", i), {n_embd, n_ff, n_expert}, false); + layer.ffn_gate_exps = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_GATE_EXPS, "weight", i), {n_embd, n_ff, n_expert}); layer.ffn_down_exps = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_DOWN_EXPS, "weight", i), {n_ff, n_embd, n_expert}); layer.ffn_up_exps = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_UP_EXPS, "weight", i), {n_embd, n_ff, n_expert});