diff options
Diffstat (limited to 'examples/main')
| -rw-r--r-- | examples/main/main.cpp | 13 |
1 files changed, 7 insertions, 6 deletions
diff --git a/examples/main/main.cpp b/examples/main/main.cpp index 711f162d..249fc2bb 100644 --- a/examples/main/main.cpp +++ b/examples/main/main.cpp @@ -246,6 +246,7 @@ int main(int argc, char ** argv) { } const bool add_bos = llama_should_add_bos_token(model); + GGML_ASSERT(llama_add_eos_token(model) != 1); LOG("add_bos: %d\n", add_bos); std::vector<llama_token> embd_inp; @@ -255,7 +256,7 @@ int main(int argc, char ** argv) { if (params.chatml) { params.prompt = "<|im_start|>system\n" + params.prompt + "<|im_end|>"; } - embd_inp = ::llama_tokenize(ctx, params.prompt, add_bos, true); + embd_inp = ::llama_tokenize(ctx, params.prompt, true, true); } else { LOG("use session tokens\n"); embd_inp = session_tokens; @@ -277,10 +278,10 @@ int main(int argc, char ** argv) { if (ctx_guidance) { LOG("cfg_negative_prompt: \"%s\"\n", log_tostr(sparams.cfg_negative_prompt)); - guidance_inp = ::llama_tokenize(ctx_guidance, sparams.cfg_negative_prompt, add_bos, true); + guidance_inp = ::llama_tokenize(ctx_guidance, sparams.cfg_negative_prompt, true, true); LOG("guidance_inp tokenized: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx_guidance, guidance_inp).c_str()); - std::vector<llama_token> original_inp = ::llama_tokenize(ctx, params.prompt, add_bos, true); + std::vector<llama_token> original_inp = ::llama_tokenize(ctx, params.prompt, true, true); LOG("original_inp tokenized: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx, original_inp).c_str()); original_prompt_len = original_inp.size(); @@ -339,14 +340,14 @@ int main(int argc, char ** argv) { } // prefix & suffix for instruct mode - const auto inp_pfx = ::llama_tokenize(ctx, "\n\n### Instruction:\n\n", add_bos, true); - const auto inp_sfx = ::llama_tokenize(ctx, "\n\n### Response:\n\n", false, true); + const auto inp_pfx = ::llama_tokenize(ctx, "\n\n### Instruction:\n\n", true, true); + const auto inp_sfx = ::llama_tokenize(ctx, "\n\n### Response:\n\n", false, true); LOG("inp_pfx: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx, inp_pfx).c_str()); LOG("inp_sfx: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx, inp_sfx).c_str()); // chatml prefix & suffix - const auto cml_pfx = ::llama_tokenize(ctx, "\n<|im_start|>user\n", add_bos, true); + const auto cml_pfx = ::llama_tokenize(ctx, "\n<|im_start|>user\n", true, true); const auto cml_sfx = ::llama_tokenize(ctx, "<|im_end|>\n<|im_start|>assistant\n", false, true); LOG("cml_pfx: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx, cml_pfx).c_str()); |