.gitignore
.pre-commit-config.yaml
LICENSE
Makefile
NOTICE
README.md
REPLACE_LIGER.md
pyproject.toml
setup.py
.github/workflows/ci.yml
.github/workflows/main-source-guard.yml
.github/workflows/publish.yml
benchmark/results/nvidia-h100-80gb-hbm3/fp8_base.json
benchmark/results/perarch/0.3.2_remove_fused_mlp.md
benchmark/results/perarch/0.4.0_fp8_frozen_base.md
benchmark/results/perarch/0.4.10_fp8_dx_backward.md
benchmark/results/perarch/0.4.11_fp8_ada_sm89.md
benchmark/results/perarch/0.4.1_lora_smallm.md
benchmark/results/perarch/0.4.2_gdn_conv_epilogue.md
benchmark/results/perarch/0.4.3_trainable_qknorm_rope.md
benchmark/results/perarch/0.4.4_arch_matrix_gdn_qknorm.md
benchmark/results/perarch/0.4.5_gdn_native_gva.md
benchmark/results/perarch/0.4.6_attn_output_gate.md
benchmark/results/perarch/0.4.7_largerdim_verify.md
benchmark/results/perarch/0.4.8_largerdim_h100_5090.md
benchmark/results/perarch/0.4.9_gdn_gated_rmsnorm.md
benchmark/results/perarch/PERARCH_PROGRESS.md
benchmark/results/perarch/flce_sweep.json
benchmark/results/perarch/sm120_cc_baseline.json
benchmark/results/perarch/sm120_cc_tuned.json
benchmark/results/perarch/sm80_cc.json
benchmark/results/perarch/sm86_cc.json
benchmark/results/perarch/sm90_cc_baseline.json
benchmark/results/perarch/sm90_cc_tuned.json
benchmark/results/perarch/sm90_flce_sweep.json
benchmark/results/perarch/nvidia-a100-sxm4-40gb/gdn_block.json
benchmark/results/perarch/nvidia-a100-sxm4-40gb/gdn_conv_gate.json
benchmark/results/perarch/nvidia-a100-sxm4-40gb/gdn_native_gva.json
benchmark/results/perarch/nvidia-a100-sxm4-40gb/largerdim_verify.json
benchmark/results/perarch/nvidia-a100-sxm4-40gb/out_gate.json
benchmark/results/perarch/nvidia-a100-sxm4-40gb/qknorm_rope_bench.json
benchmark/results/perarch/nvidia-a40/flce_chunked_oracle.json
benchmark/results/perarch/nvidia-a40/gdn_block.json
benchmark/results/perarch/nvidia-a40/gdn_conv_gate.json
benchmark/results/perarch/nvidia-a40/gdn_native_gva.json
benchmark/results/perarch/nvidia-a40/gnorm_bench.json
benchmark/results/perarch/nvidia-a40/largerdim_verify.json
benchmark/results/perarch/nvidia-a40/out_gate.json
benchmark/results/perarch/nvidia-a40/qknorm_rope_bench.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/fp8_dx_bwd.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/fp8_dx_step_bench.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/fp8_verify_h100.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/gdn_block.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/gdn_chalk_stack.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/gdn_conv_gate.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/gdn_overhead.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/gdn_profile.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/gnorm_bench.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/largerdim_verify.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/lora_bench_all.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/lora_prod.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/out_gate.json
benchmark/results/perarch/nvidia-h100-80gb-hbm3/qknorm_rope_bench.json
benchmark/results/perarch/nvidia-rtx-4090/fp8_ada_e2e.json
benchmark/results/perarch/nvidia-rtx-4090/fp8_ada_verify.json
benchmark/results/perarch/nvidia-rtx-5090/flce_chunked_oracle.json
benchmark/results/perarch/nvidia-rtx-5090/gdn_block.json
benchmark/results/perarch/nvidia-rtx-5090/gdn_conv_gate.json
benchmark/results/perarch/nvidia-rtx-5090/gdn_native_gva.json
benchmark/results/perarch/nvidia-rtx-5090/largerdim_verify.json
benchmark/results/perarch/nvidia-rtx-5090/lora_prod.json
benchmark/results/perarch/nvidia-rtx-5090/qknorm_rope_bench.json
dev/bench_all.py
dev/fmt-requirements.txt
src/chalk/__init__.py
src/chalk/utils.py
src/chalk/ops/__init__.py
src/chalk/ops/embedding.py
src/chalk/ops/flce.py
src/chalk/ops/fp8_base.py
src/chalk/ops/gdn.py
src/chalk/ops/lora.py
src/chalk/ops/qkv.py
src/chalk/ops/rmsnorm.py
src/chalk/ops/rope.py
src/chalk/ops/swiglu.py
src/chalk/transformers/__init__.py
src/chalk/transformers/apply.py
src/chalk/transformers/embedding.py
src/chalk/transformers/flce.py
src/chalk/transformers/fp8_base.py
src/chalk/transformers/gdn.py
src/chalk/transformers/lora.py
src/chalk/transformers/qkv.py
src/chalk/transformers/rmsnorm.py
src/chalk/transformers/rope.py
src/chalk/transformers/swiglu.py
src/freesolo_chalk.egg-info/PKG-INFO
src/freesolo_chalk.egg-info/SOURCES.txt
src/freesolo_chalk.egg-info/dependency_links.txt
src/freesolo_chalk.egg-info/requires.txt
src/freesolo_chalk.egg-info/top_level.txt
test/__init__.py
test/conftest.py
test/test_bench_all.py
test/test_helpers.py
test/test_package.py
test/test_smoke.py
test/test_utils.py
test/utils.py
test/ops/__init__.py
test/ops/test_embedding.py
test/ops/test_flce.py
test/ops/test_fp8_base.py
test/ops/test_gdn.py
test/ops/test_lora.py
test/ops/test_ops_contract.py
test/ops/test_qkv.py
test/ops/test_rmsnorm.py
test/ops/test_rope.py
test/ops/test_swiglu.py
test/transformers/__init__.py
test/transformers/test_apply_aggregator.py
test/transformers/test_installer_noop.py
test/transformers/test_kernel_install.py
test/transformers/test_rope_install.py
test/transformers/test_transformers_contract.py