mirror of
https://github.com/tinygrad/tinygrad.git
synced 2026-01-26 15:28:10 -05:00
* two stage cumsum in tensor.py * 2 more kernels for llama cumsum * gpt-2 and llama use fast multinomial
* two stage cumsum in tensor.py * 2 more kernels for llama cumsum * gpt-2 and llama use fast multinomial