fix perplexity after c-api refactor (#390)

* preallocate a buffer of fitting size for tokenization (utils.cpp)

* don't create a new std::string (especially here, where it's usually large)

main.cpp

@@ -85,7 +85,7 @@ void perplexity(llama_context * ctx, const gpt_params & params) {
     // Download: https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-raw-v1.zip?ref=salesforce-research
     // Run `./main --perplexity -m models/7B/ggml-model-q4_0.bin -f wiki.test.raw`
     // Output: `perplexity: 13.5106 [114/114]`
-    auto tokens = ::llama_tokenize(ctx, params.prompt.c_str(), true);
+    auto tokens = ::llama_tokenize(ctx, params.prompt, true);
 
     int count = 0;
     double nll = 0.0;

utils.cpp

@@ -146,8 +146,10 @@ std::string gpt_random_prompt(std::mt19937 & rng) {
 
 // TODO: not great allocating this every time
 std::vector<llama_token> llama_tokenize(struct llama_context * ctx, const std::string & text, bool add_bos) {
-    std::vector<llama_token> res(8096);
+    // initialize to prompt numer of chars, since n_tokens <= n_prompt_chars
+    std::vector<llama_token> res(text.size() + (int)add_bos);
     int n = llama_tokenize(ctx, text.c_str(), res.data(), res.size(), add_bos);
+    assert(n >= 0);
     res.resize(n);
 
     return res;