fix merge

Nexesenex · Nexesenex · commit e55c1df327dc · 2025-05-14T00:58:09.000+02:00
diff --git a/ggml/src/ggml-cuda/common.cuh b/ggml/src/ggml-cuda/common.cuh
@@ -281,7 +281,6 @@ static __device__ void no_device_code(
     printf("%s:%d: ERROR: CUDA kernel %s has no device code compatible with CUDA arch %d. ggml-cuda.cu was compiled for: %s\n",
            file_name, line, function_name, arch, arch_list);
 #endif // defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__)
-    __trap();
 
     GGML_UNUSED(no_device_code); // suppress unused function warning
 
diff --git a/gpttype_adapter.cpp b/gpttype_adapter.cpp
@@ -3073,7 +3073,6 @@ ModelLoadResult gpttype_load_model(const load_model_inputs inputs, FileFormat in
         }
 
         llama_ctx_params.offload_kqv = !inputs.low_vram;
-        llama_ctx_params.logits_all = false;
         model_params.use_mmap = inputs.use_mmap;
         model_params.use_mlock = inputs.use_mlock;
         model_params.n_gpu_layers = inputs.gpulayers;

Original file line number	Diff line number	Diff line change
`@@ -3073,7 +3073,6 @@ ModelLoadResult gpttype_load_model(const load_model_inputs inputs, FileFormat in`
`3073`	`3073`	`}`
`3074`	`3074`
`3075`	`3075`	`llama_ctx_params.offload_kqv = !inputs.low_vram;`
`3076`		`- llama_ctx_params.logits_all = false;`
`3077`	`3076`	`model_params.use_mmap = inputs.use_mmap;`
`3078`	`3077`	`model_params.use_mlock = inputs.use_mlock;`
`3079`	`3078`	`model_params.n_gpu_layers = inputs.gpulayers;`