compilade a1631e53f6 llama : simplify Mamba with advanced batch splits (#8526) ... * llama : advanced batch splits This includes equal-sequence-length batch splits which are useful to simplify recurrent model operators. * llama : always make recurrent state slots contiguous * ggml : simplify mamba operators * llama : fix integer signedness mixing * llama : logits_all has priority over batch->logits Otherwise, the server embeddings tests failed. This was likely an existing problem but was only detected here because of an additional assertion. * llama : apply suggestions Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * llama : fix t5 segfault * llama : fix Mamba session save and restore * llama : minor cosmetic changes * llama : rename llama_reorder_outputs to llama_output_reorder Also move it closer to llama_output_reserve. * llama : fix pooled embeddings when using batches with equal_seqs * minor : add struct members for clarity ggml-ci * llama : fix T5 segfault again * llama : fix Mamba pooled embeddings with multiple sequences Until the pooled embeddings are refactored to allow splitting across ubatches for causal embeddings, recurrent models can only process a single sequence per ubatch when calculating pooled embeddings. * llama : add llama_model_is_recurrent to simplify figuring that out This will make it easier to more cleanly support RWKV-v6 and Mamba-2. * llama : fix simple splits when the batch contains embeddings --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>		2024-08-21 17:58:11 -04:00
..
llama.h	llama : simplify Mamba with advanced batch splits (#8526)	2024-08-21 17:58:11 -04:00