| | evaluate | |
| | experimental | |
| | params | |
| | runner | |
| | source_transformation | |
| | tests | |
| | tokenizer | |
| | __init__.py | 272 B |
| | Android3_2_1B_bf16.gif | 4.8 MB |
| | Android3_2_3B_SpinQuant.gif | 7.3 MB |
| | attention.py | 17.0 KB |
| | CMakeLists.txt | 6.4 KB |
| | eval_llama_lib.py | 13.7 KB |
| | eval_llama.py | 1009 B |
| | export_llama_lib.py | 50.9 KB |
| | export_llama.py | 841 B |
| | fairseq2.py | 2.3 KB |
| | hf_download.py | 1.6 KB |
| | install_requirement_helper.py | 807 B |
| | install_requirements.sh | 663 B |
| | LICENSE | 6.8 KB |
| | llama_transformer.py | 10.2 KB |
| | llama_via_xnnpack.gif | 6.8 MB |
| | main.cpp | 3.1 KB |
| | model_args.py | 3.8 KB |
| | model.py | 16.4 KB |
| | non_cpu_backends.md | 1.3 KB |
| | norm.py | 1.5 KB |
| | README.md | 28.2 KB |
| | rope.py | 12.2 KB |
| | static_attention.py | 13.1 KB |
| | TARGETS | 8.1 KB |
| | targets.bzl | 1.0 KB |
| | test_llama_runner.sh | 478 B |
| | UTILS.md | 5.1 KB |
