9 Commits (main)

Author	SHA1	Message	Date
Jared Van Bortel	636307160e	backend: fix #includes with include-what-you-use (#2371) ... Also fix a PARENT_SCOPE warning when building the backend. Signed-off-by: Jared Van Bortel <jared@nomic.ai>	2 weeks ago
Aaron Miller	47fbc0e309	non-llama: explicitly greedy sampling for temp<=0 (#901) ... copied directly from llama.cpp - without this temp=0.0 will just scale all the logits to infinity and give bad output	1 year ago
Aaron Miller	b14953e136	sampling: remove incorrect offset for n_vocab (#900) ... no effect, but avoids a *potential* bug later if we use actualVocabSize - which is for when a model has a larger embedding tensor/# of output logits than actually trained token to allow room for adding extras in finetuning - presently all of our models have had "placeholder" tokens in the vocab so this hasn't broken anything, but if the sizes did differ we want the equivalent of `logits[actualVocabSize:]` (the start point is unchanged), not `logits[-actualVocabSize:]` (this.)	1 year ago
AT	48275d0dcc	Dlopen backend 5 (#779) ... Major change to the backend that allows for pluggable versions of llama.cpp/ggml. This was squashed merged from dlopen_backend_5 where the history is preserved.	1 year ago
Adam Treat	7f9f91ad94	Revert "New tokenizer implementation for MPT and GPT-J" ... This reverts commit bbcee1ced5.	1 year ago
Aaron Miller	bbcee1ced5	New tokenizer implementation for MPT and GPT-J ... Improves output quality by making these tokenizers more closely match the behavior of the huggingface `tokenizers` based BPE tokenizers these models were trained with. Featuring: * Fixed unicode handling (via ICU) * Fixed BPE token merge handling * Complete added vocabulary handling	1 year ago
Aaron Miller	d14936bfd6	backend: dedupe tokenizing code in mpt/gptj	1 year ago
Aaron Miller	6182026c70	backend: dedupe tokenizing code in gptj/mpt	1 year ago
Adam Treat	d918b02c29	Move the llmodel C API to new top-level directory and version it.	1 year ago