Datasets:

arm-bench
/

arm-bench-trace

Languages:

Size:

Tags:

License:

Dataset card Files Files and versions

arm-bench-trace / definitions /gemm

26.1 kB

3 contributors

History: 10 commits

awu12's picture

Tag q4_k_m defs correctness:sqnr + stable SiLU (moe)

7b9ac69 verified about 1 month ago

gemm_bf16_n1024_k2048.json

832 Bytes
Add kernel in different precision: about 1 month ago
gemm_bf16_n1408_k2048.json

838 Bytes
Add kernel in different precision: about 1 month ago
gemm_bf16_n2048_k1024.json

832 Bytes
Add kernel in different precision: about 1 month ago
gemm_bf16_n2048_k1408.json

838 Bytes
Add kernel in different precision: about 1 month ago
gemm_bf16_n2048_k2048.json

837 Bytes
Add kernel in different precision: about 1 month ago
gemm_fp32_n1000_k1280.json

785 Bytes
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_fp32_n1000_k2048.json

776 Bytes
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_fp32_n1280_k960.json

782 Bytes
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_fp32_n29_k800.json

781 Bytes
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_q4_k_m_n1024_k2048.json

1.65 kB
Tag q4_k_m defs correctness:sqnr + stable SiLU (moe) about 1 month ago
gemm_q4_k_m_n1408_k2048.json

1.66 kB
Tag q4_k_m defs correctness:sqnr + stable SiLU (moe) about 1 month ago
gemm_q4_k_m_n2048_k1024.json

1.65 kB
Tag q4_k_m defs correctness:sqnr + stable SiLU (moe) about 1 month ago
gemm_q4_k_m_n2048_k1536.json

1.84 kB
Tag q4_k_m defs correctness:sqnr + stable SiLU (moe) about 1 month ago
gemm_q4_k_m_n2048_k2048.json

1.66 kB
Tag q4_k_m defs correctness:sqnr + stable SiLU (moe) about 1 month ago
gemm_q8_0_n1024_k2048.json

1.38 kB
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_q8_0_n1408_k2048.json

1.39 kB
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_q8_0_n2048_k1024.json

1.38 kB
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_q8_0_n2048_k1408.json

1.39 kB
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_q8_0_n2048_k2048.json

1.39 kB
add baseline solution source at all neural network kernel definitions about 1 month ago
gemm_w8a8ch_n1000_k1280.json

1.14 kB
Update all w8a8ch definitions' correctness reference, aligns with what llama.cpp low-bit kernel correctness reference style: dequant input, then calculate in full precision about 1 month ago
gemm_w8a8ch_n1000_k2048.json

1.13 kB
Update all w8a8ch definitions' correctness reference, aligns with what llama.cpp low-bit kernel correctness reference style: dequant input, then calculate in full precision about 1 month ago
gemm_w8a8ch_n1280_k960.json

1.14 kB
Update all w8a8ch definitions' correctness reference, aligns with what llama.cpp low-bit kernel correctness reference style: dequant input, then calculate in full precision about 1 month ago