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/batched.swift/Sources/main.swift | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'examples/batched.swift/Sources/main.swift') diff --git a/examples/batched.swift/Sources/main.swift b/examples/batched.swift/Sources/main.swift index d75c503d..5764acb6 100644 --- a/examples/batched.swift/Sources/main.swift +++ b/examples/batched.swift/Sources/main.swift @@ -153,7 +153,7 @@ while n_cur <= n_len { // const llama_token new_token_id = llama_sample_token_greedy(ctx, &candidates_p); // is it an end of stream? -> mark the stream as finished - if new_token_id == llama_token_eos(model) || n_cur == n_len { + if llama_token_is_eog(model, new_token_id) || n_cur == n_len { i_batch[i] = -1 // print("") if n_parallel > 1 { -- cgit v1.2.3