diff options
| author | Joan Fontanals <jfontanalsmartinez@gmail.com> | 2024-05-11 09:46:09 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-05-11 10:46:09 +0300 |
| commit | b83cc3f5b303ff30c52874b2d5864dc6385ebf9f (patch) | |
| tree | 912fe3b8df9eb7b39ae7f4321f080b924628336b /examples | |
| parent | 9cb317f77e53067f7a138cc89ef7657148eae8e6 (diff) | |
llama : add Jina Embeddings architecture (#6826)
* feat: first things to do
* feat: create tensors for Jina architecture
* fix: use other tensors
* feat: embedding gets results
* fix: fix usage of ALIBI
* fix: clean prints
* fix: do some cleanup unused vars
* fix: revert changes to Makefile and CMakeLists
* fix: revert some changes
* fix: fix small detail
* fix: fix convert formatting
* fix: fix linting and editor
* feat: set proper vocab settings
* fix: JinaBertForMaskedLM registration
* feat: support q_normalization and k_normalization in Jina arch
* feat: handle gpt2 tokenizer with Jina architecture
* feat: example comments in embedding
* feat: rename Jina Bert to Jina Bert V2
* fix: add some changes as per review
* feat: proper KQ_pos for Jina embeddings
* feat: add capacity to load models ES and DE for Spanish
* llama : fix pre-tokenizers
* ggml : full ALiBi support
* ggml : update ggml_soft_max_ext() CUDA, SYCL
* ggml : ggml_flash_attn_ext() support ALiBi (CPU)
* ggml : ggml_flash_attn_ext() support ALiBi (Metal)
* ggml : fix warning
* ggml : ggml_flash_attn_ext() support ALiBi (CUDA)
ggml-ci
* minor : clean-up
* embedding : add warning about missing SEP
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Diffstat (limited to 'examples')
| -rw-r--r-- | examples/embedding/embedding.cpp | 12 |
1 files changed, 10 insertions, 2 deletions
diff --git a/examples/embedding/embedding.cpp b/examples/embedding/embedding.cpp index 6a93147d..c85a2da5 100644 --- a/examples/embedding/embedding.cpp +++ b/examples/embedding/embedding.cpp @@ -49,6 +49,12 @@ static void batch_decode(llama_context * ctx, llama_batch & batch, float * outpu } float * out = output + batch.seq_id[i][0] * n_embd; + //TODO: I would also add a parameter here to enable normalization or not. + /*fprintf(stdout, "unnormalized_embedding:"); + for (int hh = 0; hh < n_embd; hh++) { + fprintf(stdout, "%9.6f ", embd[hh]); + } + fprintf(stdout, "\n");*/ llama_embd_normalize(embd, out, n_embd); } } @@ -123,10 +129,12 @@ int main(int argc, char ** argv) { inputs.push_back(inp); } - // add SEP if not present + // check if the last token is SEP + // it should be automatically added by the tokenizer when 'tokenizer.ggml.add_eos_token' is set to 'true' for (auto & inp : inputs) { if (inp.empty() || inp.back() != llama_token_sep(model)) { - inp.push_back(llama_token_sep(model)); + fprintf(stderr, "%s: warning: last token in the prompt is not SEP\n", __func__); + fprintf(stderr, "%s: 'tokenizer.ggml.add_eos_token' should be set to 'true' in the GGUF header\n", __func__); } } |