Merge branch 'add-fh1-rebased' of https://github.com/tiiuae/llama.cpp-public into add-fh1-rebased

Author: ibrahimkhadraoui

src/llama-arch.cpp

@@ -129,7 +129,7 @@ static const std::map<llm_kv, const char *> LLM_KV_NAMES = {
     { LLM_KV_EMBEDDING_SCALE,                   "%s.embedding_scale"                   },
     { LLM_KV_TOKEN_SHIFT_COUNT,                 "%s.token_shift_count"                 },
     { LLM_KV_INTERLEAVE_MOE_LAYER_STEP,         "%s.interleave_moe_layer_step"         },
-    
+
     { LLM_KV_ATTENTION_HEAD_COUNT,                   "%s.attention.head_count"                   },
     { LLM_KV_ATTENTION_HEAD_COUNT_KV,                "%s.attention.head_count_kv"                },
     { LLM_KV_ATTENTION_MAX_ALIBI_BIAS,               "%s.attention.max_alibi_bias"               },

src/llama-hparams.cpp

@@ -74,8 +74,8 @@ uint32_t llama_hparams::n_embd_r() const {
     // TODO: maybe support other convolution strides than 1
     // NOTE: since the first column of the conv_state is shifted out each time, it's not actually needed
     // Corresponds to Mamba's conv_states size
-    
-    // check if the architecture is using d_ssm 
+
+    // check if the architecture is using d_ssm
     return (ssm_d_conv > 0 ? ssm_d_conv - 1 : 0) * (ssm_d_inner + 2*ssm_n_group*ssm_d_state);
 }
 

src/llama-model.cpp

@@ -4600,7 +4600,7 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
                     if (output == NULL) {
                         output = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, TENSOR_DUPLICATED);
                     }
-                    
+
                     for (int i = 0; i < n_layer; ++i) {
                         auto & layer = layers[i];
                         layer.attn_norm = create_tensor(tn(LLM_TENSOR_ATTN_NORM, "weight", i), {n_embd}, 0);
@@ -14738,7 +14738,7 @@ struct llm_build_falcon_h1 : public llm_graph_context {
             inpSA = ggml_add(ctx0, cur, inpSA);
             cb(cur, "layer_out", il);
 
-            if (il == n_layer - 1 && inp_out_ids) {            
+            if (il == n_layer - 1 && inp_out_ids) {
                 cur   = ggml_get_rows(ctx0,   cur, inp_out_ids);
                 inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids);
             }