.gitignore
Doxyfile
LICENSE
MANIFEST.in
README.md
config.py
env.src
setup.py
thunderkittens.cpp
assets/attn.png
assets/thunderkittens.png
demos/README.md
demos/requirements.txt
demos/assets/kittens.png
demos/based_demo/README.md
demos/based_demo/document_ie_based.py
demos/based_demo/generate_based.py
demos/based_demo/based/__init__.py
demos/based_demo/based/models/__init__.py
demos/based_demo/based/models/layer_norm.py
demos/based_demo/based/models/mixers/__init__.py
demos/based_demo/based/models/mixers/convolution.py
demos/based_demo/based/models/mixers/linear_attention.py
demos/based_demo/based/models/mixers/rotary.py
demos/based_demo/based/models/mixers/slide_attention.py
demos/based_demo/plots/benchmark-lin-attn-fwd-B16.png
demos/based_demo/plots/benchmark-lin-attn-fwd-B4.png
demos/based_demo/plots/benchmark-lin-attn-fwd-L8192.png
demos/based_demo/plots/benchmark_input1000_output1.png
demos/based_demo/plots/benchmark_input16000_output1.png
demos/based_demo/plots/benchmark_input8000_output1.png
demos/based_demo/train/src/__init__.py
demos/based_demo/train/src/generation.py
demos/based_demo/train/src/models/__init__.py
demos/based_demo/train/src/models/block.py
demos/based_demo/train/src/models/embeddings.py
demos/based_demo/train/src/models/gpt.py
demos/based_demo/train/src/models/mha.py
demos/based_demo/train/src/models/mlp.py
demos/based_demo/train/src/utils/hf.py
demos/based_demo/train/src/utils/utils.py
demos/configs/llama_3.1_8b_distill_config.yaml
demos/configs/llama_3.1_8b_finetune_config.yaml
demos/configs/llama_3.1_8b_model_config.yaml
demos/llama_demo/demo_8b.sh
demos/llama_demo/demo_llama_hf.py
demos/llama_demo/src/__init__.py
demos/llama_demo/src/model/__init__.py
demos/llama_demo/src/model/pretrained.py
demos/llama_demo/src/model/transformers_modeling_llama.py
demos/llama_demo/src/model/transformers_modeling_utils.py
demos/llama_demo/src/utils/__init__.py
demos/llama_demo/src/utils/logging.py
demos/llama_demo/src/utils/setup.py
demos/lolcats_demo/demo_8b.sh
demos/lolcats_demo/demo_lolcats_hf.py
demos/lolcats_demo/src/__init__.py
demos/lolcats_demo/src/model/__init__.py
demos/lolcats_demo/src/model/convert_model.py
demos/lolcats_demo/src/model/feature_map.py
demos/lolcats_demo/src/model/load_model.py
demos/lolcats_demo/src/model/modeling_llama.py
demos/lolcats_demo/src/model/peft.py
demos/lolcats_demo/src/model/pretrained.py
demos/lolcats_demo/src/model/rotary.py
demos/lolcats_demo/src/model/utils.py
demos/lolcats_demo/src/model/linear_attention/__init__.py
demos/lolcats_demo/src/model/linear_attention/linear_attention.py
demos/lolcats_demo/src/model/linear_attention/linear_window_attention_tk.py
demos/lolcats_demo/src/model/linear_attention/linear_window_attention_tk_gen.py
demos/lolcats_demo/src/model/linear_attention/linear_window_attention_tk_long.py
demos/lolcats_demo/src/model/linear_attention/utils.py
demos/lolcats_demo/src/utils/__init__.py
demos/lolcats_demo/src/utils/logging.py
demos/lolcats_demo/src/utils/setup.py
demos/qwen_demo/demo_8b.sh
demos/qwen_demo/demo_qwen_hf.py
demos/qwen_demo/src/__init__.py
demos/qwen_demo/src/model/__init__.py
demos/qwen_demo/src/model/pretrained.py
demos/qwen_demo/src/model/transformers_modeling_qwen.py
demos/qwen_demo/src/model/transformers_modeling_utils.py
demos/qwen_demo/src/utils/__init__.py
demos/qwen_demo/src/utils/logging.py
demos/qwen_demo/src/utils/setup.py
docs/conda_setup.md
include/kittens.cuh
include/common/base_ops.cuh
include/common/base_types.cuh
include/common/common.cuh
include/common/debug.cuh
include/common/util.cuh
include/common/pyutils/test_build_utils.py
include/common/pyutils/torch_helpers.cuh
include/ops/ops.cuh
include/ops/group/group.cuh
include/ops/group/memory/memory.cuh
include/ops/group/memory/tile/global_to_register.cuh
include/ops/group/memory/tile/global_to_shared.cuh
include/ops/group/memory/tile/shared_to_register.cuh
include/ops/group/memory/tile/tile.cuh
include/ops/group/memory/tile/complex/complex_global_to_register.cuh
include/ops/group/memory/tile/complex/complex_global_to_shared.cuh
include/ops/group/memory/tile/complex/complex_shared_to_register.cuh
include/ops/group/memory/util/util.cuh
include/ops/group/memory/vec/global_to_register.cuh
include/ops/group/memory/vec/global_to_shared.cuh
include/ops/group/memory/vec/shared_to_register.cuh
include/ops/group/memory/vec/vec.cuh
include/ops/group/shared/shared.cuh
include/ops/group/shared/tile/conversions.cuh
include/ops/group/shared/tile/maps.cuh
include/ops/group/shared/tile/reductions.cuh
include/ops/group/shared/tile/tile.cuh
include/ops/group/shared/vec/conversions.cuh
include/ops/group/shared/vec/maps.cuh
include/ops/group/shared/vec/vec.cuh
include/ops/group/wgmma/wgmma.cuh
include/ops/group/wgmma/base/64x112.impl
include/ops/group/wgmma/base/64x128.impl
include/ops/group/wgmma/base/64x144.impl
include/ops/group/wgmma/base/64x16.impl
include/ops/group/wgmma/base/64x160.impl
include/ops/group/wgmma/base/64x176.impl
include/ops/group/wgmma/base/64x192.impl
include/ops/group/wgmma/base/64x208.impl
include/ops/group/wgmma/base/64x224.impl
include/ops/group/wgmma/base/64x240.impl
include/ops/group/wgmma/base/64x256.impl
include/ops/group/wgmma/base/64x32.impl
include/ops/group/wgmma/base/64x48.impl
include/ops/group/wgmma/base/64x64.impl
include/ops/group/wgmma/base/64x80.impl
include/ops/group/wgmma/base/64x96.impl
include/ops/group/wgmma/base/base.cuh
include/ops/warp/warp.cuh
include/ops/warp/memory/memory.cuh
include/ops/warp/memory/tile/global_to_register.cuh
include/ops/warp/memory/tile/global_to_shared.cuh
include/ops/warp/memory/tile/shared_to_register.cuh
include/ops/warp/memory/tile/tile.cuh
include/ops/warp/memory/tile/tma.cuh
include/ops/warp/memory/tile/complex/complex_global_to_register.cuh
include/ops/warp/memory/tile/complex/complex_global_to_shared.cuh
include/ops/warp/memory/tile/complex/complex_shared_to_register.cuh
include/ops/warp/memory/util/tma.cuh
include/ops/warp/memory/util/util.cuh
include/ops/warp/memory/vec/global_to_register.cuh
include/ops/warp/memory/vec/global_to_shared.cuh
include/ops/warp/memory/vec/shared_to_register.cuh
include/ops/warp/memory/vec/tma.cuh
include/ops/warp/memory/vec/vec.cuh
include/ops/warp/register/register.cuh
include/ops/warp/register/tile/conversions.cuh
include/ops/warp/register/tile/maps.cuh
include/ops/warp/register/tile/mma.cuh
include/ops/warp/register/tile/reductions.cuh
include/ops/warp/register/tile/tile.cuh
include/ops/warp/register/tile/complex/complex_conversions.cuh
include/ops/warp/register/tile/complex/complex_maps.cuh
include/ops/warp/register/tile/complex/complex_mma.cuh
include/ops/warp/register/vec/conversions.cuh
include/ops/warp/register/vec/maps.cuh
include/ops/warp/register/vec/reductions.cuh
include/ops/warp/register/vec/vec.cuh
include/ops/warp/shared/shared.cuh
include/ops/warp/shared/tile/conversions.cuh
include/ops/warp/shared/tile/maps.cuh
include/ops/warp/shared/tile/reductions.cuh
include/ops/warp/shared/tile/tile.cuh
include/ops/warp/shared/vec/conversions.cuh
include/ops/warp/shared/vec/maps.cuh
include/ops/warp/shared/vec/reductions.cuh
include/ops/warp/shared/vec/vec.cuh
include/types/types.cuh
include/types/global/cgl.cuh
include/types/global/gl.cuh
include/types/global/global.cuh
include/types/global/tma.cuh
include/types/global/util.cuh
include/types/register/crt.cuh
include/types/register/crv.cuh
include/types/register/register.cuh
include/types/register/rt.cuh
include/types/register/rt_base.cuh
include/types/register/rt_layout.cuh
include/types/register/rv.cuh
include/types/register/rv_layout.cuh
include/types/shared/cst.cuh
include/types/shared/csv.cuh
include/types/shared/shared.cuh
include/types/shared/st.cuh
include/types/shared/sv.cuh
kernels/attn/demo/4090.cu
kernels/attn/demo/4090_harness.impl
kernels/attn/demo/Makefile
kernels/attn/demo/README.md
kernels/attn/demo/gentests.py
kernels/attn/demo/h100_lcf.cu
kernels/attn/demo/h100_lcf_harness.impl
kernels/attn/h100/Makefile
kernels/attn/h100/attn
kernels/attn/h100/attn.d
kernels/attn/h100/check.py
kernels/attn/h100/gentests.py
kernels/attn/h100/h100.cu
kernels/attn/h100/h100_bench.py
kernels/attn/h100/h100_check.py
kernels/attn/h100/hammer_bwd.py
kernels/attn/h100/hammer_fwd.py
kernels/attn/h100/hammer_fwd_bwd.py
kernels/attn/h100/harness.impl
kernels/layernorm/non_pc/Makefile
kernels/layernorm/non_pc/gentests.py
kernels/layernorm/non_pc/harness.impl
kernels/layernorm/non_pc/layer_norm.cu
kernels/linear/Makefile
kernels/linear/linear.cu
kernels/matmul/H100/Makefile
kernels/matmul/H100/matmul
kernels/matmul/H100/matmul.cu
kernels/ttt/.gitignore
kernels/ttt/Makefile
kernels/ttt/harness.cuh
kernels/ttt/matching.py
kernels/ttt/torch_harness.cuh
kernels/ttt/ttt.cu
kernels/ttt/triton_comps/linear_forward.py
kernels/ttt/triton_comps/mlp_forward_split.py
kernels/ttt_backward/matching.py
kernels/ttt_backward/test.cpp
kernels/ttt_backward/ttt.cu
kernels/ttt_backward/triton_comps/linear_backward.py
kernels/ttt_backward/triton_comps/mlp_backward_split.py
kernels/ttt_tp/.gitignore
kernels/ttt_tp/Makefile
kernels/ttt_tp/ttt_tp.cu
kernels/ttt_tp/ttt_tp.d
prototype/prototype.cuh
prototype/common/common.cuh
prototype/common/templates.cuh
prototype/common/util.cuh
prototype/lcf/lcf.cuh
prototype/lcf/templates.cuh
prototype/lcsf/lcsf.cuh
prototype/lcsf/templates.cuh
test_time_training.egg-info/PKG-INFO
test_time_training.egg-info/SOURCES.txt
test_time_training.egg-info/dependency_links.txt
test_time_training.egg-info/requires.txt
test_time_training.egg-info/top_level.txt
thunderkittens.egg-info/PKG-INFO
thunderkittens.egg-info/SOURCES.txt
thunderkittens.egg-info/dependency_links.txt
thunderkittens.egg-info/top_level.txt
ttt.egg-info/PKG-INFO
ttt.egg-info/SOURCES.txt
ttt.egg-info/dependency_links.txt
ttt.egg-info/top_level.txt
ttt_mlp.egg-info/PKG-INFO
ttt_mlp.egg-info/SOURCES.txt
ttt_mlp.egg-info/dependency_links.txt
ttt_mlp.egg-info/requires.txt
ttt_mlp.egg-info/top_level.txt
ttt_tk.egg-info/PKG-INFO
ttt_tk.egg-info/SOURCES.txt
ttt_tk.egg-info/dependency_links.txt
ttt_tk.egg-info/top_level.txt