Merge commit 'd4e0d95cf581f50c9a21d06eaecae2dd580076bd' into concedo_experimental

# Conflicts:
#	.github/workflows/build.yml
#	common/CMakeLists.txt
#	ggml/src/CMakeLists.txt
#	ggml/src/ggml-opencl/CMakeLists.txt
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-rpc/ggml-rpc.cpp
#	scripts/sync-ggml.last
#	tests/CMakeLists.txt

tools/server/server.cpp

@@ -233,6 +233,7 @@ struct server_task {
         slot_params defaults;
         defaults.sampling    = params_base.sampling;
         defaults.speculative = params_base.speculative;
+        defaults.n_keep      = params_base.n_keep;
 
         // enabling this will output extra debug information in the HTTP responses from the server
         params.verbose           = params_base.verbosity > 9;
@@ -2060,6 +2061,7 @@ struct server_context {
             SLT_INF(slot, "new slot n_ctx_slot = %d\n", slot.n_ctx);
 
             slot.params.sampling = params_base.sampling;
+            slot.params.n_keep = params_base.n_keep;
 
             slot.callback_on_release = [this](int) {
                 queue_tasks.pop_deferred_task();
@@ -3556,9 +3558,6 @@ struct server_context {
                 const llama_tokens & cached_text_tokens = slot.cache_tokens.get_text_tokens();
                 llama_tokens draft = common_speculative_gen_draft(slot.spec, params_spec, cached_text_tokens, id);
 
-                // keep track of total number of tokens generated in the draft
-                slot.n_draft_total += draft.size();
-
                 // ignore small drafts
                 if (slot.params.speculative.n_min > (int) draft.size()) {
                     SLT_DBG(slot, "ignoring small draft: %d < %d\n", (int) draft.size(), slot.params.speculative.n_min);
@@ -3566,6 +3565,9 @@ struct server_context {
                     continue;
                 }
 
+                // keep track of total number of drafted tokens tested
+                slot.n_draft_total += draft.size();
+
                 // construct the speculation batch
                 common_batch_clear(slot.batch_spec);
                 common_batch_add  (slot.batch_spec, id, slot.n_past, { slot.id }, true);
@@ -3584,7 +3586,7 @@ struct server_context {
                 slot.n_past    += ids.size();
                 slot.n_decoded += ids.size();
 
-                // update how many tokens out of draft was accepted
+                // update how many tokens out of those tested were accepted
                 slot.n_draft_accepted += ids.size() - 1;
 
                 slot.cache_tokens.push_back(id);

tools/server/webui/src/index.scss

@@ -41,6 +41,10 @@ html {
   max-width: 900px;
 }
 
+.chat-bubble {
+  @apply break-words;
+}
+
 .chat-bubble-base-300 {
   --tw-bg-opacity: 1;
   --tw-text-opacity: 1;