Francis LamandGitHub 9d142430cb Add option in llama.py to quantize weights to int8 at runtime (#1289) ...

* Add option in llama.py to quantize weights to int8 at runtime

Also added lm-eval to external

* Add support for llama-2 evaluation

2023-07-24 17:22:38 -07:00

..

external_copy_benchmark.py

good changes from the M1 Tensor Core project (#730)

2023-03-29 05:11:02 +04:00

external_hlb_cifar.py

Fix naming conflict with huggingface datasets (#1161)

2023-07-07 10:43:44 -07:00

external_llama_eval.py

Add option in llama.py to quantize weights to int8 at runtime (#1289)

2023-07-24 17:22:38 -07:00

external_metal_uaf.py

fix metal uaf (#964)

2023-06-09 21:28:06 -07:00

external_multi_gpu.py

helpers.py: improved test coverage + exception handling (#1165)

2023-07-07 10:26:05 -07:00

external_osx_profiling.py

multi cl_queue (#762)

2023-05-03 12:15:28 -07:00

external_test_gpu_ast.py

move to shapetracker.py

2023-03-11 07:50:07 -08:00

external_test_image.py

Add RDNA3 assembler UOps.CAST partial support + other fixes/improvements (#1012)

2023-06-20 11:34:10 -07:00

external_test_llvm.py

fix up external tests

2023-03-06 06:52:28 -08:00

external_test_onnx_backend.py

exclude a few more onnx tests

2023-06-19 08:51:29 -07:00

external_test_opt.py

flake8 (#1323)

2023-07-24 11:19:58 -04:00

external_test_optim.py

Add LAMB optimizer (#821)

2023-05-28 15:09:05 -07:00

external_test_speed_llama.py

Add LLaMA-2 support (#1284)

2023-07-24 17:12:02 -04:00

external_test_yolo.py

test cleanups

2023-03-03 06:36:06 -08:00

external_test_yolov8.py

CI < 5 minutes (#1252)

2023-07-23 13:00:56 -07:00

fuzz_shapetracker.py

Add Fuzz Test symbolic / shapetracker to CI. (#1278)

2023-07-19 09:05:45 -07:00

fuzz_symbolic.py

symbolic stride (#1326)

2023-07-23 12:41:22 -07:00

graph_batchnorm.py

Refactor nn.optim (#1091)

2023-07-02 15:07:30 -07:00