Skip to content

Actions: google/gemma.cpp

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
2,500+ workflow runs
2,500+ workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Use the current SIMD target's width for the transposed KV cache build #8121: Pull request #1027 opened by Magnushst
40s Magnushst:fix/flash-attention-nf
Magnushst:fix/flash-attention-nf
40s
Use the current SIMD target's width for the transposed KV cache GitHub Actions Scan #9: Pull request #1027 opened by Magnushst
1m 7s
1m 7s
Use the current SIMD target's width for the transposed KV cache build #8120: Pull request #1026 opened by Magnushst
1m 13s Magnushst:fix/flash-attention-nf
Magnushst:fix/flash-attention-nf
1m 13s
Avoids repeated heap allocations/deallocations of layout.total_bytes per attention subtask and kv_out_mem per token step by storing worker_workspaces and kv_out_mem on AttentionActivations / AttentionActivationsPtrs. This eliminates ~20M minor page faults and improves decode throughput by +18% to +26% on AMD Turin. build #8119: Pull request #1025 opened by copybara-service Bot
26m 9s test_979178228
test_979178228
26m 9s
Avoids repeated heap allocations/deallocations of `layout.total_bytes… build #8118: Commit 5c1be13 pushed by copybara-service Bot
27m 32s test_979178228
test_979178228
27m 32s
Decompose Gelu and FastGelu into GeluCdf and FastGeluCdf, reformulati… build #8117: Commit 5321954 pushed by copybara-service Bot
20m 0s dev
dev
20m 0s
Decompose Gelu and FastGelu into GeluCdf and FastGeluCdf, reformulating the build #8116: Pull request #1024 synchronize by copybara-service Bot
20m 45s test_978577053
test_978577053
20m 45s
Decompose Gelu and FastGelu into GeluCdf and FastGeluCdf, reformulati… build #8115: Commit 5321954 pushed by copybara-service Bot
21m 15s test_978577053
test_978577053
21m 15s
Decompose Gelu and FastGelu into GeluCdf and FastGeluCdf, reformulating the build #8114: Pull request #1024 synchronize by copybara-service Bot
6m 34s test_978577053
test_978577053
6m 34s
Decompose Gelu and FastGelu into GeluCdf and FastGeluCdf, reformulati… build #8113: Commit b1a69c6 pushed by copybara-service Bot
6m 42s test_978577053
test_978577053
6m 42s
Decompose Gelu and FastGelu into GeluCdf and FastGeluCdf, reformulating the build #8112: Pull request #1024 opened by copybara-service Bot
26m 3s test_978577053
test_978577053
26m 3s
Decompose Gelu and FastGelu into GeluCdf and FastGeluCdf, reformulati… build #8111: Commit 05a5b90 pushed by copybara-service Bot
25m 43s test_978577053
test_978577053
25m 43s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize un… build #8110: Commit 824b1b8 pushed by copybara-service Bot
25m 55s dev
dev
25m 55s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize unsigned-by-signed (uint8 * int8) matrix multiplication for SV accumulation, improving quantization precision. Add SIMD emulation fallback using SumOfMulQuadAccumulate and TableLookupBytes. build #8109: Pull request #1013 synchronize by copybara-service Bot
27m 30s test_976294440
test_976294440
27m 30s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize un… build #8108: Commit 824b1b8 pushed by copybara-service Bot
26m 3s test_976294440
test_976294440
26m 3s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize unsigned-by-signed (uint8 * int8) matrix multiplication for SV accumulation, improving quantization precision. Add SIMD emulation fallback using SumOfMulQuadAccumulate and TableLookupBytes. build #8107: Pull request #1013 synchronize by copybara-service Bot
25m 0s test_976294440
test_976294440
25m 0s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize un… build #8106: Commit 030a7d1 pushed by copybara-service Bot
26m 27s test_976294440
test_976294440
26m 27s
Reduce local KV memory in BF16 Flash attention build #8105: Pull request #1020 synchronize by Mikyx-1
Action required Mikyx-1:fix/local-kv-rings-dev
Mikyx-1:fix/local-kv-rings-dev
Action required
Quantize softmax probabilities S into [0, 255] uint8_t and utilize unsigned-by-signed (uint8 * int8) matrix multiplication for SV accumulation, improving quantization precision. Add SIMD emulation fallback using SumOfMulQuadAccumulate and TableLookupBytes. build #8104: Pull request #1013 synchronize by copybara-service Bot
25m 46s test_976294440
test_976294440
25m 46s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize un… build #8103: Commit 2768324 pushed by copybara-service Bot
25m 52s test_976294440
test_976294440
25m 52s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize unsigned-by-signed (uint8 * int8) matrix multiplication for SV accumulation, improving quantization precision. Add SIMD emulation fallback using SumOfMulQuadAccumulate and TableLookupBytes. build #8102: Pull request #1013 synchronize by copybara-service Bot
26m 27s test_976294440
test_976294440
26m 27s
Quantize softmax probabilities S into [0, 255] uint8_t and utilize un… build #8101: Commit e85ec12 pushed by copybara-service Bot
30m 13s test_976294440
test_976294440
30m 13s
In TileFlashAttentionSVBlockBF16, 256-bit InterleaveLower and Interle… build #8100: Commit 2f19ffb pushed by copybara-service Bot
24m 52s dev
dev
24m 52s
In TileFlashAttentionSVBlockBF16, 256-bit InterleaveLower and InterleaveUpper build #8099: Pull request #1023 synchronize by copybara-service Bot
27m 13s test_977780932
test_977780932
27m 13s
In TileFlashAttentionSVBlockBF16, 256-bit InterleaveLower and Interle… build #8098: Commit 2f19ffb pushed by copybara-service Bot
27m 4s test_977780932
test_977780932
27m 4s