katsu560 a84ab1da8d tests : fix quantize perf (#1990) пре 2 година
..
CMakeLists.txt f954edda93 ggml : implement backward pass for llama + small training-llama-from-scratch example (#1360) пре 2 година
test-double-float.c 436e561931 all : be more strict about converting float to double (#458) пре 2 година
test-grad0.c 65bdd52a86 tests : sync test-grad0 from ggml пре 2 година
test-opt.c f954edda93 ggml : implement backward pass for llama + small training-llama-from-scratch example (#1360) пре 2 година
test-quantize-fns.cpp 9cbf50c041 build : fix and ignore MSVC warnings (#1889) пре 2 година
test-quantize-perf.cpp a84ab1da8d tests : fix quantize perf (#1990) пре 2 година
test-sampling.cpp b061ba9e2a llama : fix top-p sampling to match the canonical definition (#1953) пре 2 година
test-tokenizer-0.cpp 527b6fba1d llama : make model stateless and context stateful (llama_state) (#1797) пре 2 година