Skip to content

Actions: am17an/llama.cpp

Actions

Check Pre-Tokenizer Hashes

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
pre-tokenizer-hashes.yml
19 workflow runs
19 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

CUDA: mul_mat_id for mmf for bs <= 64 for f16 and bs <= 32 for f32 Check Pre-Tokenizer Hashes #19: Commit 3e67208 pushed by am17an
1m 3s cuda_mmf_mmid
cuda_mmf_mmid
1m 3s
Fix bug: fix tie breakers Check Pre-Tokenizer Hashes #18: Commit a275f10 pushed by am17an
54s cuda_topk_moe
cuda_topk_moe
54s
Refactor into ggml_cuda_should_use_topk_moe Check Pre-Tokenizer Hashes #17: Commit 613b6c3 pushed by am17an
50s cuda_topk_moe
cuda_topk_moe
50s
CUDA: add topk kernel for MoE Check Pre-Tokenizer Hashes #16: Commit 2789c9e pushed by am17an
50s cuda_topk_moe
cuda_topk_moe
50s
Address review comments Check Pre-Tokenizer Hashes #15: Commit bcf81ed pushed by am17an
55s llada_moe
llada_moe
55s
Address review comments Check Pre-Tokenizer Hashes #14: Commit 2f3ab34 pushed by am17an
57s llada_moe
llada_moe
57s
Address review comments Check Pre-Tokenizer Hashes #13: Commit f109414 pushed by am17an
47s llada_moe
llada_moe
47s
Add convert_hf_to_gguf_update change Check Pre-Tokenizer Hashes #12: Commit 347b769 pushed by am17an
45s llada_moe
llada_moe
45s
Add LLaDA-7b-MoE diffusion model Check Pre-Tokenizer Hashes #11: Commit 2c2d1e6 pushed by am17an
51s llada_moe
llada_moe
51s
Add LLaDA-7b-moe diffusion model Check Pre-Tokenizer Hashes #10: Commit 08b76a7 pushed by am17an
45s llada_moe
llada_moe
45s
Remove tempalte from n_expert_used Check Pre-Tokenizer Hashes #9: Commit bf08ea5 pushed by am17an
52s mmf_opt_cuda
mmf_opt_cuda
52s
fixup Check Pre-Tokenizer Hashes #8: Commit 823e1a1 pushed by am17an
51s mmf_opt_cuda
mmf_opt_cuda
51s
Divide mmf by ncols_dst Check Pre-Tokenizer Hashes #7: Commit e6528a3 pushed by am17an
32s mul_mat_id_for_mmf
mul_mat_id_for_mmf
32s
CANN: Resolve soft_max precision issue (#15730) Check Pre-Tokenizer Hashes #6: Commit 9961d24 pushed by am17an
11m 7s master
master
11m 7s
CANN: FIx compiler warnings (#15661) Check Pre-Tokenizer Hashes #5: Commit ef47691 pushed by am17an
29s master
master
29s
CUDA: fix bug in rms_norm fusion Check Pre-Tokenizer Hashes #4: Commit 5521b2e pushed by am17an
34s fix_rms_norm_add
fix_rms_norm_add
34s
format Check Pre-Tokenizer Hashes #3: Commit 49cf98e pushed by am17an
28s rms_norm_fused_add
rms_norm_fused_add
28s
CUDA cmake: add -lineinfo for easier debug Check Pre-Tokenizer Hashes #2: Commit e737ac8 pushed by am17an
32s cuda_cmake_add_debug
cuda_cmake_add_debug
32s
CUDA: add attention sinks for tile and wmma Check Pre-Tokenizer Hashes #1: Commit 4946c19 pushed by am17an
29s cuda_fattn_tile_wmma
cuda_fattn_tile_wmma
29s