Update README.md

Requeue build

Update README.md

Slim the Space to runtime files only — tests and docs live in DaisyChainAI/DaisyChain-Train web/

Upload README.md with huggingface_hub

README: front matter + full body (front matter must never be dropped)

Restore Space README front matter (clobbered by bulk upload)

Dispatch-optimized backward (bit-identical) + IEEE-754 oracle + oracle mutation tests + TEST_RESULTS.md

Exact kernel gates: verify variants emitted from the same WGSL source restore the int32 oracle on the kernels training actually runs; bit-exact epilogue mirror removes the tolerance entirely; shape sweep; dead exactly-gated kernel deleted. Continuous verification: random-cell audit at live shapes + canonical cross-device kernel probe (the weight hash structurally cannot detect kernel disagreement). New test_gates.js mutation-tests the gate itself.

Update README.md

Split-K f32 backward (CUTLASS ex.06) + gather-fused attention kernels (ex.36/52): compute step 821ms -> 420ms

Update README.md

CUTLASS-style: block-scaled INT8 quantization, batched attention GEMM (one dispatch for all heads), fused dequant+ReLU epilogue

Update README.md

Leader failover, 5-retry reconnect then drop, dataset picker, pretraining framing

Room-first lobby (Create/Join, host approval), group training settings sliders

DaisyChain-Web demo: P2P browser training through verified INT8 units

initial commit