From 29ae62d2ae163e2b68aa0ad3bf2ab4636de0c957 Mon Sep 17 00:00:00 2001 From: Georgi Gerganov Date: Mon, 4 Mar 2024 22:31:20 +0200 Subject: llama : fix embeddings (#5796) * llama : fix embeddings ggml-ci * llama : do not use KV cache for non-causal models ggml-ci * embeddings : fix llama_batch_init arg * llama : add pooling switch * llama : distinguish token vs sequence embeddings ggml-ci * llama : assert pooling tensor * llama : simplify causal mask condition ggml-ci * llama : assert input batch with pooling enabled * readme : update API changes list --- common/common.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'common/common.cpp') diff --git a/common/common.cpp b/common/common.cpp index 036a9813..c244db64 100644 --- a/common/common.cpp +++ b/common/common.cpp @@ -1292,7 +1292,7 @@ struct llama_context_params llama_context_params_from_gpt_params(const gpt_param cparams.n_threads_batch = params.n_threads_batch == -1 ? params.n_threads : params.n_threads_batch; cparams.seed = params.seed; cparams.logits_all = params.logits_all; - cparams.embedding = params.embedding; + cparams.embeddings = params.embedding; cparams.rope_scaling_type = params.rope_scaling_type; cparams.rope_freq_base = params.rope_freq_base; cparams.rope_freq_scale = params.rope_freq_scale; -- cgit v1.2.3