Name | Size | Last Modified |
|---|---|---|
| amp_C_frontend.cpp | 4.1 KB | |
| compat.h | 140 bytes | |
| flatten_unflatten.cpp | 584 bytes | |
| layer_norm_cuda.cpp | 6.5 KB | |
| layer_norm_cuda_kernel.cu | 25 KB | |
| mlp.cpp | 4.6 KB | |
| mlp_cuda.cu | 52.3 KB | |
| multi_tensor_adagrad.cu | 3 KB | |
| multi_tensor_adam.cu | 4.5 KB | |
| multi_tensor_apply.cuh | 4.7 KB | |
| multi_tensor_axpby_kernel.cu | 4.7 KB | |
| multi_tensor_l2norm_kernel.cu | 12.8 KB | |
| multi_tensor_lamb.cu | 12.6 KB | |
| multi_tensor_lamb_stage_1.cu | 4.4 KB | |
| multi_tensor_lamb_stage_2.cu | 3.4 KB | |
| multi_tensor_novograd.cu | 5.1 KB | |
| multi_tensor_scale_kernel.cu | 4.1 KB | |
| multi_tensor_sgd_kernel.cu | 8.1 KB | |
| syncbn.cpp | 6.2 KB | |
| type_shim.h | 4.6 KB | |
| welford.cu | 54.6 KB |
