From b97bc3966e852adb626c90be64fd48282800f504 Mon Sep 17 00:00:00 2001 From: Pedro Cuenca Date: Sun, 21 Apr 2024 13:50:41 +0200 Subject: llama : support Llama 3 HF conversion (#6745) * Support Llama 3 conversion The tokenizer is BPE. * style * Accept suggestion Co-authored-by: Sourab Mangrulkar <13534540+pacman100@users.noreply.github.com> * llama : add llama_token_is_eog() ggml-ci * llama : auto-detect more EOT tokens when missing in KV data * convert : replacing EOS token is a hack * llama : fix codegemma EOT token + add TODOs * llama : fix model type string for 8B model --------- Co-authored-by: Sourab Mangrulkar <13534540+pacman100@users.noreply.github.com> Co-authored-by: Georgi Gerganov --- examples/main/main.cpp | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'examples/main/main.cpp') diff --git a/examples/main/main.cpp b/examples/main/main.cpp index 249fc2bb..1180734b 100644 --- a/examples/main/main.cpp +++ b/examples/main/main.cpp @@ -795,8 +795,8 @@ int main(int argc, char ** argv) { } } - // deal with end of text token in interactive mode - if (llama_sampling_last(ctx_sampling) == llama_token_eos(model)) { + // deal with end of generation tokens in interactive mode + if (llama_token_is_eog(model, llama_sampling_last(ctx_sampling))) { LOG("found EOS token\n"); if (params.interactive) { @@ -920,8 +920,8 @@ int main(int argc, char ** argv) { } } - // end of text token - if (!embd.empty() && embd.back() == llama_token_eos(model) && !(params.instruct || params.interactive || params.chatml)) { + // end of generation + if (!embd.empty() && llama_token_is_eog(model, embd.back()) && !(params.instruct || params.interactive || params.chatml)) { LOG_TEE(" [end of text]\n"); break; } -- cgit v1.2.3