cleanup

crates/llama-cpp-bindings/src/engine.cc

@@ -34,8 +34,6 @@ struct Request {
   std::string generated_text;
 
 
-  std::vector<llama_token> past_tokens;
-
   std::vector<llama_token> find_candidate_pred_tokens(size_t ngram_size = 3, size_t n_pred_tokens = 8) {
     if (past_tokens.size() < ngram_size) return {};
     std::vector<llama_token> ngram(past_tokens.begin() + past_tokens.size() - ngram_size, past_tokens.end());
@@ -51,6 +49,15 @@ struct Request {
   size_t n_past() {
     return past_tokens.size();
   }
+
+  void step(llama_token next_token, size_t n_dropped) {
+    past_tokens.insert(past_tokens.end(), tokens.begin(), tokens.begin() + tokens.size() - n_dropped);
+    tokens.clear();
+    tokens.push_back(next_token);
+  }
+
+ private:
+  std::vector<llama_token> past_tokens;
 };
 
 
@@ -235,15 +242,12 @@ class TextInferenceEngineImpl : public TextInferenceEngine {
         }
 
         llama_token next_token = -1;
-        size_t n_tokens = request.tokens.size() - request.n_draft - 1;
-        request.past_tokens.insert(request.past_tokens.end(), request.tokens.begin(), request.tokens.begin() + n_tokens);
-
+        int k = -request.n_draft;
         // FIXME: ensure batching logic always put i_batch - request.n_draft in this batch.
-        for (int k = -request.n_draft; k < 1; ++k) {
+        for (k = -request.n_draft; k < 1; ++k) {
           auto logits = llama_get_logits_ith(ctx, i_batch + k);
           next_token = std::distance(logits, std::max_element(logits, logits + n_vocab));
 
-          request.past_tokens.push_back(next_token);
           const auto token_str = llama_token_to_piece(ctx, next_token);
           request.generated_text += token_str;
 
@@ -286,15 +290,13 @@ class TextInferenceEngineImpl : public TextInferenceEngine {
             break;
           }
 
-          if ((k < 0 && next_token != request.tokens[request.tokens.size() + k])) {
+          if ((k == 0) || ((k < 0 && next_token != request.tokens[request.tokens.size() + k]))) {
+            request.step(next_token, -k);
             llama_kv_cache_seq_rm(ctx_.get(), request.id, request.n_past(), -1);
             break;
           }
         }
 
-        request.tokens.clear();
-        request.tokens.push_back(next_token);
-
         auto draft_tokens = request.find_candidate_pred_tokens();
         request.n_draft = draft_tokens.size();
         request.tokens.insert(request.tokens.end(), draft_tokens.begin(), draft_tokens.end());