MCPcopy Create free account

hub / github.com/HazyResearch/HipKittens / functions

Functions1,340 in github.com/HazyResearch/HipKittens

⨍ Functions 1,340◇ Types & classes 190↳ Endpoints 1
Methodgrid
kernels/attn/gqa_backwards/archive/GQA_bkwd_prep.cpp:184
Methodgrid
kernels/attn/gqa_causal_backwards/attn_bkwd_prep.cpp:199
Methodgrid
training/llama/csrc/attn_bkwd_prep.cpp:199
Functionhip_check
kernels/gemm/bf16fp32/gfx1250/harness.h:28
Methodinit_defaults
(self)
training/llama/train/datamodules/base.py:171
Functioninit_randint
(shape, low, high, dtype, device)
kernels/gemm/bf16fp32/utils.py:3
Methodinit_weights
(self)
training/llama/llama/models/seq_common.py:276
Functioninit_zero
(shape, dtype, device)
kernels/gemm/bf16fp32/utils.py:12
Functioninterpolate_pos_embedding
(state_dict, out_seqlen, pos_embedding_name='model.pos_encoder.pe', interleave=False)
training/llama/train/utils/checkpoint.py:41
Functionkeep
(conf)
analysis/baselines/attn/triton_baseline_v01.py:92
Functionlayer_norm_fn
( x, weight, bias, residual=None, x1=None, weight1=None, bias1=None, eps=1e-6,
training/llama/llama/ops/triton/layer_norm.py:859
Functionlayer_norm_linear_fn
( x, norm_weight, norm_bias, linear_weight, linear_bias, residual=None, eps=1e-6,
training/llama/llama/ops/triton/layer_norm.py:1063
Functionlayer_norm_ref
( x, weight, bias, residual=None, x1=None, weight1=None, bias1=None, eps=1e-6,
training/llama/llama/ops/triton/layer_norm.py:19
Functionlayernorm_tk
__launch_bounds__(NUM_THREADS, 2)
analysis/layernorm/mi350x/kernel_4096.cpp:85
Functionlayernorm_tk
__launch_bounds__(NUM_THREADS, 2)
analysis/layernorm/mi350x/kernel_1024.cpp:86
Functionlayernorm_tk
__launch_bounds__(NUM_THREADS, 2)
analysis/layernorm/mi350x/kernel_2048.cpp:85
Functionlayernorm_tk
__launch_bounds__(NUM_THREADS, 2)
analysis/layernorm/mi350x/kernel_16384.cpp:85
Functionlayernorm_tk
__launch_bounds__(NUM_THREADS, 2)
analysis/layernorm/mi350x/kernel_8192.cpp:85
Functionlayernorm_tk
__launch_bounds__(NUM_THREADS, 2)
kernels/layernorm/kernel.cpp:85
Functionload_async_shared_to_register
analysis/bf16_gemm/mi325x/utils.cpp:168
Functionload_async_shared_to_register
kernels/gemm/bf16fp32/archive/utils.cpp:154
Functionload_global_to_registers
analysis/bf16_gemm/mi325x/utils.cpp:43
Functionload_global_to_registers
kernels/gemm/bf16fp32/archive/utils.cpp:59
Functionload_global_to_shared_direct
kernels/gemm/bf16fp32/archive/utils.cpp:224
Functionload_global_to_shared_direct_with_swizzled_offsets
kernels/gemm/bf16fp32/archive/utils.cpp:316
Functionload_global_vec_new
analysis/bf16_gemm/mi325x/utils.cpp:5
Functionload_one
kernels/gemm/fp8fp32/FP8_4wave/utils.cpp:51
Functionload_shared_vec_async_offset
analysis/bf16_gemm/mi325x/utils.cpp:135
Functionload_shared_vec_b128
kernels/gemm/bf16fp32/archive/utils.cpp:354
Functionload_shuffled
analysis/attn/bkwd/benchmark/attn_bkwd_prep.cpp:48
Functionload_shuffled
kernels/attn/gqa_backwards/attn_bkwd_prep.cpp:48
Functionload_shuffled
kernels/attn/gqa_backwards/archive/GQA_bkwd_prep.cpp:35
Functionload_shuffled
kernels/attn/gqa_causal_backwards/attn_bkwd_prep.cpp:48
Functionload_shuffled
training/llama/csrc/attn_bkwd_prep.cpp:48
Functionload_st_to_rt
kernels/gemm/fp8fp32/FP8_8wave/utils.cpp:11
Functionload_st_to_rt
kernels/torch_scaled/utils.cpp:11
Methodload_state_dict
(self, state_dict)
training/llama/train/datamodules/fault_tolerant_sampler.py:29
Methodload_state_dict
(self, state_dict)
training/llama/train/datamodules/fault_tolerant_sampler.py:75
Functionmain
kernels/gemm/bf16fp32/gfx1250/harness.h:56
Functionmake_builder
(out_file, impl, vocab_size=None)
training/llama/train/datamodules/datasets/indexed_dataset.py:53
Functionmake_dataset
(path, impl, skip_warmup=False)
training/llama/train/datamodules/datasets/indexed_dataset.py:62
Functionmask_vec2_imm
analysis/attn/bkwd/benchmark/attn_fwd_causal.cpp:74
Functionmask_vec2_imm
analysis/attn/fwd/benchmark/attn_fwd_causal.cpp:95
Functionmask_vec2_imm
analysis/attn/fwd/benchmark/attn_fwd_causal_d64.cpp:95
Functionmask_vec2_imm
kernels/attn/gqa_causal_backwards/attn_fwd_causal.cpp:74
Functionmask_vec2_imm
kernels/attn/gqa_causal/kernel.cpp:95
Functionmask_vec2_imm
kernels/attn/gqa_causal/kernel_d64.cpp:95
Functionmask_vec2_imm
training/llama/csrc/attn_fwd_causal.cpp:74
Functionmatmul_kernel
Kernel for computing the matmul C = A x B. A has shape (M, K), B has shape (K, N) and C has shape (M, N)
analysis/baselines/gemm/triton_gemm_v01.py:242
Functionmax_fn
(x, y)
analysis/baselines/attn/triton_baseline_v02.py:145
Methodmerge_file_
(self, another_file)
training/llama/train/datamodules/datasets/indexed_dataset.py:310
Methodmerge_file_
(self, another_file)
training/llama/train/datamodules/datasets/indexed_dataset.py:590
Functionmicro_tk
distributed-kernels/bf16_gemm/kernel.cpp:90
Functionmicro_tk
analysis/paper_experiments/phases/ds_write_b64/kernel.cpp:19
Functionmicro_tk
analysis/paper_experiments/phases/ds_read_b128/kernel.cpp:19
Functionmicro_tk
analysis/paper_experiments/phases/ds_read_b64/kernel.cpp:19
Functionmicro_tk
analysis/paper_experiments/phases/ds_read_b96/kernel.cpp:19
Functionmicro_tk
analysis/paper_experiments/grid_micro/kernel_8192_w0.cpp:38
Functionmicro_tk
analysis/paper_experiments/grid_micro/kernel_8192_w32.cpp:38
Functionmicro_tk
analysis/paper_experiments/grid_micro/kernel_8192_w2.cpp:38
Functionmicro_tk
analysis/paper_experiments/grid_micro/kernel_8192_w8.cpp:38
Functionmicro_tk
analysis/paper_experiments/grid_micro/kernel_8192_w4.cpp:38
Functionmicro_tk
analysis/bf16_gemm/mi325x/kernel_4096.cpp:35
Functionmicro_tk
analysis/bf16_gemm/mi325x/kernel_1024.cpp:35
Functionmicro_tk
analysis/bf16_gemm/mi325x/kernel_2048.cpp:35
Functionmicro_tk
analysis/bf16_gemm/mi325x/kernel_16384.cpp:35
Functionmicro_tk
analysis/bf16_gemm/mi325x/kernel_8192.cpp:35
Functionmicro_tk
analysis/bf16_gemm/mi350x/kernel.cpp:39
Functionmicro_tk
kernels/gemm/bf16fp32/256_256_64_32_with32x16.cpp:35
Functionmicro_tk
kernels/gemm/bf16fp32/256_256_64_32_with16x32.cpp:39
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_04_2stage_12c4p.cpp:36
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_09_async.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_03_3stage_8c4p.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_02_2stage_8c4p.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/timing/micro_01_syncthreads.cpp:29
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_06_2stage_8c4p_64x96.cpp:38
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_06_2stage_8c4p_96x64.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_07_2stage_8c4p_nblock8.cpp:35
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_08_2stage_4c4p.cpp:39
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_08_async.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_08_4stage_4c4p.cpp:38
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_05_2stage_16c2p.cpp:36
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_04_2stage_12c4p.cpp:38
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_03_3stage_8c4p.cpp:38
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_02_2stage_8c4p.cpp:38
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_05_async.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/archive/micro_06_async.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/archive/micro_03_3stage_8c4p_async.cpp:38
Functionmicro_tk
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/archive/micro_07_async.cpp:37
Functionmicro_tk
kernels/gemm/bf16fp32/micros/hint_based/kernel.cpp:40
Functionmicro_tk
kernels/gemm/bf16fp32/micros/192x256/kernel.cpp:40
Functionmicro_tk
kernels/gemm/bf16fp32/micros/192x256/kernelv2.cpp:40
Functionmicro_tk
kernels/gemm/bf16fp32/archive/256_256_64_32/256_256_64_32.cpp:39
Functionmicro_tk
kernels/gemm/bf16fp32/archive/256_256_64_16/256_256_64_16.cpp:39
Functionmicro_tk
kernels/gemm/bf16fp32/archive/micros/test_accum_layout/kernel.cpp:27
Functionmicro_tk
kernels/gemm/bf16fp32/archive/micros/64_16/kernel.cpp:28
Functionmicro_tk
kernels/gemm/bf16fp32/archive/micros/32_16/kernel.cpp:28
Functionmicro_tk
kernels/gemm/bf16fp32/archive/micros/256_64/kernel.cpp:38
Functionmicro_tk
kernels/gemm/bf16fp32/archive/micros/test_load_to_lds/kernel.cpp:27
Functionmicro_tk
kernels/gemm/bf16fp32/archive/micros/transpose_matmul/256_256_64_16.cpp:35
← previousnext →1,101–1,200 of 1,340, ranked by callers