238 kB

Ctrl+K

1 contributor

History: 19 commits

ticketguy

CogMemBench paper for Kaggle + GPU memory reduction experiments

d7a2e27 verified 3 minutes ago

.gitattributes

1.52 kB
initial commit 2 days ago
README.md

734 Bytes
Update ML Intern artifact metadata 2 days ago
bench_full.py

11.1 kB
Full GPU benchmark including FP32 quant quality test about 22 hours ago
bench_train_only.py

7.3 kB
Training-only GPU benchmark (skip FP32 quant quality — already proven on CPU) about 22 hours ago
benchmark_gpu_fixed.py

10.9 kB
GPU benchmark script (fixed for PyTorch 2.11) 2 days ago
figquant_gpu_fixed.py

3.83 kB
FigQuant GPU training test (with dtype fix) about 2 hours ago
final_gpu_test.py

9.08 kB
Fix FigQuant GPU benchmark (use figcache mode) + test engine conversion about 14 hours ago
fix_all.py

7.8 kB
Fix lowram backward dtype bug + AGPL licenses + rerun GPU test about 2 hours ago
fix_compile.py

5.86 kB
Fix engine compilation errors about 2 hours ago
kaggle_paper_and_memory.py

11.2 kB
CogMemBench paper for Kaggle + GPU memory reduction experiments 3 minutes ago
lila_engine_phase1.py

27.3 kB
Lila Engine Phase 1: Foundation — matmul kernel + model loader + token generation about 21 hours ago
lila_engine_phase1b.py

14.2 kB
Engine Phase 1b: AVX2 vectorized matmul + RMSNorm kernels about 21 hours ago
lila_engine_phase2.py

19.1 kB
Engine Phase 2: Full transformer forward pass + tokenizer + attention about 17 hours ago
lila_engine_phase3.py

28 kB
Engine Phase 3: Complete format converter + BPE tokenizer + kernel wiring about 17 hours ago
lila_engine_plan.py

20 kB
Lila inference engine build plan + fix LilaCore about 22 hours ago
lila_restructure.py

32.3 kB
Lila restructure script 2 days ago
memory_and_cogmem.py

9.31 kB
GPU memory experiment + CogMemBench on real model about 1 hour ago
test_engine_compile.py

3.13 kB
Test engine compilation about 2 hours ago
update_docs.py

15.5 kB
Update paper + README with final GPU results, fix Colab, research GPU memory reduction about 1 hour ago