diff options
| author | John <78893154+cmp-nct@users.noreply.github.com> | 2024-01-18 23:12:15 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-01-19 00:12:15 +0200 |
| commit | 57e2a7a52a819883f40dada8a2edc24ecf48186b (patch) | |
| tree | 52663a84fbfcb73b36231fff3c6b4a9ba0a1773e /llama.cpp | |
| parent | 9b6ea4263ab45e02ff905bf7a29dc143ca1facc3 (diff) | |
llama : fix falcon arch for tied output embeddings (#4978)
* falcon arch fix for tied output embeddings
* Update llama.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Update llama.cpp
* Update llama.cpp
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Update llama.cpp
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Diffstat (limited to 'llama.cpp')
| -rw-r--r-- | llama.cpp | 7 |
1 files changed, 6 insertions, 1 deletions
@@ -3438,7 +3438,12 @@ static bool llm_load_tensors( { model.output_norm = ml.create_tensor(ctx_output, tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd}); model.output_norm_b = ml.create_tensor(ctx_output, tn(LLM_TENSOR_OUTPUT_NORM, "bias"), {n_embd}); - model.output = ml.create_tensor(ctx_output_split, tn(LLM_TENSOR_OUTPUT, "weight"), {n_embd, n_vocab}); + if (gguf_find_tensor(ml.ctx_gguf, tn(LLM_TENSOR_OUTPUT, "weight").c_str()) >= 0) { + model.output = ml.create_tensor(ctx_output_split, tn(LLM_TENSOR_OUTPUT, "weight"), {n_embd, n_vocab}); + } else { + model.output = ml.create_tensor(ctx_output_split, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}); // needs to be on GPU + ml.n_created--; // artificial tensor + } } for (int i = 0; i < n_layer; ++i) { |