From 5be6c803fa5378f62a1590f3ad8c6b64c7c0c2ce Mon Sep 17 00:00:00 2001 From: Marcus Dunn <51931484+MarcusDunn@users.noreply.github.com> Date: Mon, 23 Oct 2023 12:40:03 -0700 Subject: llama : remove token functions with `context` args in favor of `model` (#3720) * added `llama_model_token_*` variants to all the `llama_token_*` functions. * added `LLAMA_API` * formatting Co-authored-by: Georgi Gerganov * removed old `llama_token` functions * changed 3 more functions to take in model - `llama_token_get_text` - `llama_token_get_score` - `llama_token_get_type` * added back docs * fixed main.cpp * changed token functions to use new model variants * changed token functions to use new model variants --------- Co-authored-by: Georgi Gerganov --- examples/perplexity/perplexity.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'examples/perplexity/perplexity.cpp') diff --git a/examples/perplexity/perplexity.cpp b/examples/perplexity/perplexity.cpp index 7d0038bd..3c2542e8 100644 --- a/examples/perplexity/perplexity.cpp +++ b/examples/perplexity/perplexity.cpp @@ -227,7 +227,7 @@ static results_perplexity perplexity_v2(llama_context * ctx, const gpt_params & // add BOS token for the first batch of each chunk if (add_bos && j == 0) { - tokens[batch_start] = llama_token_bos(ctx); + tokens[batch_start] = llama_token_bos(llama_get_model(ctx)); } const auto batch_logits = llama_get_logits(ctx); @@ -350,7 +350,7 @@ static results_perplexity perplexity(llama_context * ctx, const gpt_params & par // add BOS token for the first batch of each chunk if (add_bos && j == 0) { - tokens[batch_start] = llama_token_bos(ctx); + tokens[batch_start] = llama_token_bos(llama_get_model(ctx)); } if (llama_decode(ctx, llama_batch_get_one(tokens.data() + batch_start, batch_size, j * n_batch, 0))) { -- cgit v1.2.3