llama: update vendored code to 081b29bd

api/types.go

@@ -225,7 +225,6 @@ type Options struct {
 	Mirostat         int      `json:"mirostat,omitempty"`
 	MirostatTau      float32  `json:"mirostat_tau,omitempty"`
 	MirostatEta      float32  `json:"mirostat_eta,omitempty"`
-	PenalizeNewline  bool     `json:"penalize_newline,omitempty"`
 	Stop             []string `json:"stop,omitempty"`
 }
 
@@ -602,7 +601,6 @@ func DefaultOptions() Options {
 		Mirostat:         0,
 		MirostatTau:      5.0,
 		MirostatEta:      0.1,
-		PenalizeNewline:  true,
 		Seed:             -1,
 
 		Runner: Runner{

cmd/interactive_test.go

@@ -63,17 +63,15 @@ func TestModelfileBuilder(t *testing.T) {
 			{Role: "assistant", Content: "Yes it is true, I am half horse, half shark."},
 		},
 		Options: map[string]any{
-			"temperature":      0.9,
-			"seed":             42,
-			"penalize_newline": false,
-			"stop":             []string{"hi", "there"},
+			"temperature": 0.9,
+			"seed":        42,
+			"stop":        []string{"hi", "there"},
 		},
 	}
 
 	t.Run("model", func(t *testing.T) {
 		expect := `FROM hork
 SYSTEM You are part horse and part shark, but all hork. Do horklike things
-PARAMETER penalize_newline false
 PARAMETER seed 42
 PARAMETER stop hi
 PARAMETER stop there
@@ -92,7 +90,6 @@ MESSAGE assistant Yes it is true, I am half horse, half shark.
 		opts.ParentModel = "horseshark"
 		expect := `FROM horseshark
 SYSTEM You are part horse and part shark, but all hork. Do horklike things
-PARAMETER penalize_newline false
 PARAMETER seed 42
 PARAMETER stop hi
 PARAMETER stop there

docs/api.md

@@ -396,17 +396,13 @@ curl http://localhost:11434/api/generate -d '{
     "mirostat": 1,
     "mirostat_tau": 0.8,
     "mirostat_eta": 0.6,
-    "penalize_newline": true,
     "stop": ["\n", "user:"],
     "numa": false,
     "num_ctx": 1024,
     "num_batch": 2,
     "num_gpu": 1,
     "main_gpu": 0,
-    "low_vram": false,
-    "vocab_only": false,
     "use_mmap": true,
-    "use_mlock": false,
     "num_thread": 8
   }
 }'

llama/amx.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/amx.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/clip.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/clip.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/common.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *
@@ -966,6 +966,25 @@ struct common_init_result common_init_from_params(common_params & params) {
         params.sampling.ignore_eos = false;
     }
 
+    if (params.sampling.ignore_eos) {
+        for (llama_token i = 0; i < llama_n_vocab(model); i++) {
+            if (llama_token_is_eog(model, i)) {
+                LOG_INF("%s: added %s logit bias = %f\n", __func__, common_token_to_piece(lctx, i).c_str(), -INFINITY);
+                params.sampling.logit_bias.push_back({i, -INFINITY});
+            }
+        }
+    }
+
+    if (params.sampling.penalty_last_n == -1) {
+        LOG_INF("%s: setting penalty_last_n to ctx_size = %d\n", __func__, llama_n_ctx(lctx));
+        params.sampling.penalty_last_n = llama_n_ctx(lctx);
+    }
+
+    if (params.sampling.dry_penalty_last_n == -1) {
+        LOG_INF("%s: setting dry_penalty_last_n to ctx_size = %d\n", __func__, llama_n_ctx(lctx));
+        params.sampling.dry_penalty_last_n = llama_n_ctx(lctx);
+    }
+
     if (params.warmup) {
         LOG_WRN("%s: warming up the model with an empty run - please wait ... (--no-warmup to disable)\n", __func__);
 

llama/common.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *
@@ -121,6 +121,7 @@ enum common_sampler_type {
     COMMON_SAMPLER_TYPE_TEMPERATURE = 7,
     COMMON_SAMPLER_TYPE_XTC         = 8,
     COMMON_SAMPLER_TYPE_INFILL      = 9,
+    COMMON_SAMPLER_TYPE_PENALTIES   = 10,
 };
 
 // dimensionality reduction methods, used by cvector-generator
@@ -156,7 +157,6 @@ struct common_params_sampling {
     int32_t mirostat           = 0;     // 0 = disabled, 1 = mirostat, 2 = mirostat 2.0
     float   mirostat_tau       = 5.00f; // target entropy
     float   mirostat_eta       = 0.10f; // learning rate
-    bool    penalize_nl        = false; // consider newlines as a repeatable token
     bool    ignore_eos         = false;
     bool    no_perf            = false; // disable performance metrics
     bool    timing_per_token   = false;
@@ -165,6 +165,7 @@ struct common_params_sampling {
 
 
     std::vector<enum common_sampler_type> samplers = {
+        COMMON_SAMPLER_TYPE_PENALTIES,
         COMMON_SAMPLER_TYPE_DRY,
         COMMON_SAMPLER_TYPE_TOP_K,
         COMMON_SAMPLER_TYPE_TYPICAL_P,
@@ -219,11 +220,13 @@ struct common_params {
     float   defrag_thold          =  0.1f; // KV cache defragmentation threshold
 
     // offload params
-    std::vector<ggml_backend_dev_t> devices;         // devices to use for offloading
-    int32_t n_gpu_layers                    =    -1; // number of layers to store in VRAM (-1 - use default)
-    int32_t main_gpu                        =     0; // the GPU that is used for scratch and small tensors
-    float   tensor_split[128]               =   {0}; // how split tensors should be distributed across GPUs
-    enum llama_split_mode        split_mode = LLAMA_SPLIT_MODE_LAYER; // how to split the model across GPUs
+    std::vector<ggml_backend_dev_t> devices; // devices to use for offloading
+
+    int32_t n_gpu_layers      = -1;  // number of layers to store in VRAM (-1 - use default)
+    int32_t main_gpu          = 0;   // the GPU that is used for scratch and small tensors
+    float   tensor_split[128] = {0}; // how split tensors should be distributed across GPUs
+
+    enum llama_split_mode split_mode = LLAMA_SPLIT_MODE_LAYER; // how to split the model across GPUs
 
     struct cpu_params cpuparams;
     struct cpu_params cpuparams_batch;

llama/ggml-alloc.c

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *
@@ -560,7 +560,6 @@ static void ggml_gallocr_allocate_node(ggml_gallocr_t galloc, struct ggml_tensor
         size_t offset = ggml_dyn_tallocr_alloc(alloc, size, node);
         hn->buffer_id = buffer_id;
         hn->offset = offset;
-        return;
     }
 }
 

llama/ggml-alloc.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-backend-impl.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-backend-reg.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-backend.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-backend.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-blas.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-blas.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-common.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpp.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu-aarch64.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu-aarch64.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu-impl.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu-quants.c

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu-quants.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu-traits.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu-traits.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu.c

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cpu.cpp

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *
@@ -419,8 +419,11 @@ static bool ggml_backend_cpu_device_supports_op(ggml_backend_dev_t dev, const st
     switch (op->op) {
         case GGML_OP_CPY:
             return
+                op->type != GGML_TYPE_IQ3_XXS &&
+                op->type != GGML_TYPE_IQ3_S   &&
                 op->type != GGML_TYPE_IQ2_XXS &&
                 op->type != GGML_TYPE_IQ2_XS  &&
+                op->type != GGML_TYPE_IQ2_S   &&
                 op->type != GGML_TYPE_IQ1_S   &&
                 op->type != GGML_TYPE_IQ1_M; // missing type_traits.from_float
         case GGML_OP_MUL_MAT:

llama/ggml-cpu.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda.h

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/acc.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/acc.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/arange.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/arange.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/argmax.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/argmax.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/argsort.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/argsort.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/binbcast.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/binbcast.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/clamp.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/clamp.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/common.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/concat.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/concat.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/conv-transpose-1d.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/conv-transpose-1d.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/convert.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/convert.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/count-equal.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/count-equal.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/cpy.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/cpy.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/cross-entropy-loss.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/cross-entropy-loss.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/dequantize.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/diagmask.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/diagmask.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-common.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-tile-f16.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-tile-f16.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-tile-f32.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-tile-f32.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-vec-f16.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-vec-f32.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn-wmma-f16.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/fattn.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/getrows.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/getrows.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/ggml-cuda.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/im2col.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/im2col.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/mma.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/mmq.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/mmq.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/mmv.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/mmv.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/mmvq.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/mmvq.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/norm.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/norm.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/opt-step-adamw.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/opt-step-adamw.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/out-prod.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/out-prod.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/pad.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/pad.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/pool2d.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/pool2d.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/quantize.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/quantize.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/rope.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/rope.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/scale.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/scale.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/softmax.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/softmax.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/sum.cu

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *

llama/ggml-cuda/sum.cuh

@@ -1,5 +1,5 @@
 /**
- * llama.cpp - commit ba1cb19cdd0d92e012e0f6e009e0620f854b6afd - do not edit this file
+ * llama.cpp - commit 081b29bd2a3d91e7772e3910ce223dd63b8d7d26 - do not edit this file
  *
  * MIT License
  *