Jong-hun Shin 4934d49274

Support GPT-NeoX Models without attention biases (#2301)

2023-12-30 11:42:04 -05:00

..

layers

Remove Sampler copy stream (#2209)

2023-12-20 00:04:33 -08:00

models

Support GPT-NeoX Models without attention biases (#2301)

2023-12-30 11:42:04 -05:00

parallel_utils

Remove dependency on CuPy (#2152)

2023-12-17 01:49:07 -08:00

__init__.py

Refactor Worker & InputMetadata (#1843)

2023-11-29 22:16:37 -08:00

input_metadata.py

Optimize model execution with CUDA graph (#1926)

2023-12-16 21:12:08 -08:00

model_loader.py

Implement lazy model loader (#2044)

2023-12-12 22:21:45 -08:00

sampling_metadata.py

Make sampler less blocking (#1889)

2023-12-17 23:03:49 +08:00

utils.py

TP/quantization/weight loading refactor part 2 - Refactor quantized linear logic and extend quantization support to all models (#1622)

2023-11-15 22:50:41 -08:00

weight_utils.py

[Minor] Fix a typo in .pt weight support (#2160)

2023-12-17 10:12:44 -08:00