MCPcopy Create free account

hub / github.com/HazyResearch/HipKittens / functions

Functions1,340 in github.com/HazyResearch/HipKittens

Methodblock
kernels/attn/gqa_backwards/archive/GQA_bkwd_prep.cpp:185
Methodblock
kernels/attn/gqa_causal_backwards/attn_bkwd_prep.cpp:200
Methodblock
training/llama/csrc/attn_bkwd_prep.cpp:200
Functionblockdiag_to_dense_mlp_bert
(state_dict)
training/llama/train/utils/checkpoint.py:32
Functionceil_div
kernels/gemm/bf16fp32/archive/micros/256_64/kernel.cpp:24
Functioncluster_mfma_count
kernels/gemm/bf16fp32/micros/hint_based/schedule_utils.cpp:26
Methodcompute
Compute the Perplexity. Returns: Perplexity
training/llama/train/metrics/perplexity.py:65
Functioncompute_buffer_loads
kernels/gemm/bf16fp32/micros/hint_based/schedule_utils.cpp:36
Functioncompute_ds_reads
kernels/gemm/bf16fp32/micros/hint_based/schedule_utils.cpp:45
Methodconfigure_optimizers
(self)
training/llama/train/tasks/seq.py:134
Functioncreate_doc_idx
(sizes)
training/llama/train/datamodules/datasets/indexed_dataset.py:128
Functioncross_entropy_bwd_kernel
( dlogits_ptr, # data ptrs dloss_ptr, logits_ptr, lse_ptr, labels_ptr, smoothing,
training/llama/llama/ops/triton/cross_entropy.py:98
Functioncross_entropy_fwd_kernel
( loss_ptr, # data ptrs lse_ptr, z_loss_ptr, logits_ptr, labels_ptr, smoothing, l
training/llama/llama/ops/triton/cross_entropy.py:26
Functiondataset_exists
(path, impl)
training/llama/train/datamodules/datasets/indexed_dataset.py:84
Functiondeprecated
(cls_or_func)
training/llama/train/datamodules/base.py:10
Functiondispatch
kernels/gemm/bf16fp32/gfx1250/gemm_tdm_arrive.cpp:160
Functiondispatch
kernels/gemm/bf16fp32/gfx1250/gemm_double_buf.cpp:64
Functiondispatch
kernels/gemm/bf16fp32/gfx1250/gemm_naive.cpp:64
Functiondispatch
kernels/gemm/bf16fp32/gfx1250/gemm_split_bar.cpp:68
Functiondispatch
kernels/gemm/bf16fp32/gfx1250/gemm_expert.cpp:82
Functiondispatch
kernels/gemm/bf16fp32/gfx1250/gemm_segment.cpp:68
Functiondispatch_bwd_combined
analysis/attn/bkwd/benchmark/attn_bkwd_causal_HBN.cpp:3355
Functiondispatch_bwd_combined
analysis/attn/bkwd/benchmark/attn_bkwd_non_causal_8_warps.cpp:845
Functiondispatch_bwd_combined
analysis/attn/bkwd/benchmark/attn_bkwd_causal_HNB.cpp:3355
Functiondispatch_bwd_combined
analysis/attn/bkwd/benchmark/attn_bkwd_non_causal_4_warps.cpp:1000
Functiondispatch_bwd_combined
analysis/attn/bkwd/benchmark/attn_bkwd_non_causal_HBN.cpp:3164
Functiondispatch_bwd_combined
analysis/attn/bkwd/benchmark/attn_bkwd_non_causal_HNB.cpp:3164
Functiondispatch_bwd_combined
kernels/attn/gqa_backwards/attn_bkwd_non_causal.cpp:3164
Functiondispatch_bwd_combined
kernels/attn/gqa_backwards/archive/GQA_bkwd_asm.cpp:848
Functiondispatch_bwd_combined
kernels/attn/gqa_backwards/archive/GQA_bkwd_4warps.cpp:987
Functiondispatch_bwd_combined
kernels/attn/gqa_backwards/archive/GQA_bkwd_8warps.cpp:832
Functiondispatch_bwd_combined
kernels/attn/gqa_causal_backwards/attn_bkwd_causal.cpp:3355
Functiondispatch_bwd_combined
training/llama/csrc/attn_bkwd_causal_HBN.cpp:3355
Functiondispatch_bwd_combined
training/llama/csrc/attn_bkwd_causal_HNB.cpp:3355
Functiondispatch_dq_shuffle
analysis/attn/bkwd/benchmark/attn_bkwd_prep.cpp:220
Functiondispatch_dq_shuffle
kernels/attn/gqa_backwards/attn_bkwd_prep.cpp:220
Functiondispatch_dq_shuffle
kernels/attn/gqa_backwards/archive/GQA_bkwd_prep.cpp:205
Functiondispatch_dq_shuffle
kernels/attn/gqa_causal_backwards/attn_bkwd_prep.cpp:220
Functiondispatch_dq_shuffle
training/llama/csrc/attn_bkwd_prep.cpp:220
Functiondispatch_fwd
analysis/attn/bkwd/benchmark/attn_fwd_causal.cpp:625
Functiondispatch_fwd
analysis/attn/bkwd/benchmark/attn_fwd_non_causal.cpp:510
Functiondispatch_fwd
kernels/attn/gqa_backwards/attn_fwd_non_causal.cpp:510
Functiondispatch_fwd
kernels/attn/gqa_backwards/archive/GQA_fwd.cpp:453
Functiondispatch_fwd
kernels/attn/gqa_causal_backwards/attn_fwd_causal.cpp:625
Functiondispatch_fwd
training/llama/csrc/attn_fwd_causal.cpp:625
Functiondispatch_micro
distributed-kernels/bf16_gemm/kernel.cpp:247
Functiondispatch_micro
analysis/paper_experiments/phases/ds_write_b64/kernel.cpp:46
Functiondispatch_micro
analysis/paper_experiments/phases/ds_read_b128/kernel.cpp:47
Functiondispatch_micro
analysis/paper_experiments/phases/ds_read_b64/kernel.cpp:47
Functiondispatch_micro
analysis/paper_experiments/phases/ds_read_b96/kernel.cpp:47
Functiondispatch_micro
analysis/paper_experiments/grid_micro/kernel_8192_w0.cpp:140
Functiondispatch_micro
analysis/paper_experiments/grid_micro/kernel_8192_w32.cpp:158
Functiondispatch_micro
analysis/paper_experiments/grid_micro/kernel_8192_w2.cpp:158
Functiondispatch_micro
analysis/paper_experiments/grid_micro/kernel_8192_w8.cpp:158
Functiondispatch_micro
analysis/paper_experiments/grid_micro/kernel_8192_w4.cpp:158
Functiondispatch_micro
analysis/layernorm/mi350x/kernel_4096.cpp:130
Functiondispatch_micro
analysis/layernorm/mi350x/kernel_1024.cpp:131
Functiondispatch_micro
analysis/layernorm/mi350x/kernel_2048.cpp:130
Functiondispatch_micro
analysis/layernorm/mi350x/kernel_16384.cpp:130
Functiondispatch_micro
analysis/layernorm/mi350x/kernel_8192.cpp:130
Functiondispatch_micro
analysis/attn/fwd/benchmark/attn_fwd_causal.cpp:696
Functiondispatch_micro
analysis/attn/fwd/benchmark/attn_fwd_causal_d64.cpp:696
Functiondispatch_micro
analysis/attn/fwd/benchmark/attn_fwd_non_causal.cpp:576
Functiondispatch_micro
analysis/attn/fwd/benchmark/attn_fwd_non_causal_d64.cpp:577
Functiondispatch_micro
analysis/bf16_gemm/mi325x/kernel_4096.cpp:227
Functiondispatch_micro
analysis/bf16_gemm/mi325x/kernel_1024.cpp:227
Functiondispatch_micro
analysis/bf16_gemm/mi325x/kernel_2048.cpp:227
Functiondispatch_micro
analysis/bf16_gemm/mi325x/kernel_16384.cpp:227
Functiondispatch_micro
analysis/bf16_gemm/mi325x/kernel_8192.cpp:227
Functiondispatch_micro
analysis/bf16_gemm/mi350x/kernel.cpp:329
Functiondispatch_micro
kernels/gemm/bf16fp32/256_256_64_32_with32x16.cpp:154
Functiondispatch_micro
kernels/gemm/bf16fp32/256_256_64_32_with16x32.cpp:329
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_04_2stage_12c4p.cpp:161
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_09_async.cpp:213
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_03_3stage_8c4p.cpp:181
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/micro_02_2stage_8c4p.cpp:156
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/timing/micro_01_syncthreads.cpp:49
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_06_2stage_8c4p_64x96.cpp:143
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_06_2stage_8c4p_96x64.cpp:132
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_07_2stage_8c4p_nblock8.cpp:146
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_08_2stage_4c4p.cpp:176
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_08_async.cpp:182
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_08_4stage_4c4p.cpp:168
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/32x16/archive/micro_05_2stage_16c2p.cpp:162
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_04_2stage_12c4p.cpp:209
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_03_3stage_8c4p.cpp:249
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_02_2stage_8c4p.cpp:208
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/micro_05_async.cpp:236
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/archive/micro_06_async.cpp:251
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/archive/micro_03_3stage_8c4p_async.cpp:243
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/producer_consumer/16x32/archive/micro_07_async.cpp:252
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/hint_based/kernel.cpp:157
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/192x256/kernel.cpp:269
Functiondispatch_micro
kernels/gemm/bf16fp32/micros/192x256/kernelv2.cpp:260
Functiondispatch_micro
kernels/gemm/bf16fp32/archive/256_256_64_32/256_256_64_32.cpp:164
Functiondispatch_micro
kernels/gemm/bf16fp32/archive/256_256_64_16/256_256_64_16.cpp:213
Functiondispatch_micro
kernels/gemm/bf16fp32/archive/micros/test_accum_layout/kernel.cpp:40
Functiondispatch_micro
kernels/gemm/bf16fp32/archive/micros/64_16/kernel.cpp:66
Functiondispatch_micro
kernels/gemm/bf16fp32/archive/micros/32_16/kernel.cpp:66
Functiondispatch_micro
kernels/gemm/bf16fp32/archive/micros/256_64/kernel.cpp:131
← previousnext →901–1,000 of 1,340, ranked by callers