fixed the chat template check condition

Author: Lpzhan931

gguf-py/gguf/constants.py

@@ -794,7 +794,7 @@ class MODEL_TENSOR(IntEnum):
     MODEL_ARCH.APERTUS:          "apertus",
     MODEL_ARCH.MINIMAXM2:        "minimax-m2",
     MODEL_ARCH.COGVLM:           "cogvlm",
-    MODEL_ARCH.PANGU_EMBED:      "pangu_embedded",
+    MODEL_ARCH.PANGU_EMBED:      "pangu-embedded",
 }
 
 VISION_PROJECTOR_TYPE_NAMES: dict[VISION_PROJECTOR_TYPE, str] = {

src/llama-arch.cpp

@@ -107,7 +107,7 @@ static const std::map<llm_arch, const char *> LLM_ARCH_NAMES = {
     { LLM_ARCH_APERTUS,          "apertus"          },
     { LLM_ARCH_MINIMAX_M2,       "minimax-m2"       },
     { LLM_ARCH_COGVLM,           "cogvlm"           },
-    { LLM_ARCH_PANGU_EMBED,      "pangu_embedded"   },
+    { LLM_ARCH_PANGU_EMBED,      "pangu-embedded"   },
     { LLM_ARCH_UNKNOWN,          "(unknown)"        },
 };
 

src/llama-chat.cpp

@@ -73,7 +73,7 @@ static const std::map<std::string, llm_chat_template> LLM_CHAT_TEMPLATES = {
     { "kimi-k2",           LLM_CHAT_TEMPLATE_KIMI_K2           },
     { "seed_oss",          LLM_CHAT_TEMPLATE_SEED_OSS          },
     { "grok-2",            LLM_CHAT_TEMPLATE_GROK_2            },
-    { "pangu_embedded",    LLM_CHAT_TEMPLATE_PANGU_EMBED       },
+    { "pangu-embedded",    LLM_CHAT_TEMPLATE_PANGU_EMBED       },
 };
 
 llm_chat_template llm_chat_template_from_str(const std::string & name) {
@@ -214,7 +214,7 @@ llm_chat_template llm_chat_detect_template(const std::string & tmpl) {
         return LLM_CHAT_TEMPLATE_SEED_OSS;
     } else if (tmpl_contains("'Assistant: '  + message['content'] + '<|separator|>")) {
         return LLM_CHAT_TEMPLATE_GROK_2;
-    } else if (tmpl_contains("[unused9]") && tmpl_contains("message['content'] + '[unused10]'")) {
+    } else if (tmpl_contains(LU8("[unused9]系统：[unused10]")) && tmpl_contains("message['content'] + '[unused10]'")) {
         return LLM_CHAT_TEMPLATE_PANGU_EMBED;
     }
     return LLM_CHAT_TEMPLATE_UNKNOWN;

src/llama-model.cpp

@@ -6281,7 +6281,7 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
                     output      = create_tensor(tn(LLM_TENSOR_OUTPUT,      "weight"), {n_embd, n_vocab}, TENSOR_NOT_REQUIRED);
 
                     // if output is NULL, init from the input tok embed
-                    if(output == NULL){
+                    if (output == NULL){
                         output = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, TENSOR_DUPLICATED);
                     }