mrq 85b9dd47c1 ugh		2025-03-19 13:31:50 -05:00
..
arch	more tweaks to the new implementation (properly trim the len stuff to save some params, decoder to d_ffn expansion to 2 to maybe also make it faster, etc.)	2025-03-18 19:34:37 -05:00
__init__.py	segregated experimental changes into its own streamlined file to avoid breaking the existing model, and it can pivot to the cleaned up code if it actually works (nothing is working)	2025-02-26 21:26:13 -06:00
ar_nar_v2.py	re-enable these	2025-03-18 20:59:33 -05:00
ar_nar.py	a gorillionth time's the charm (aka: the encoder/decoder pill is a tough pill to swallow)	2025-02-28 17:56:50 -06:00
base_v2.py	ugh	2025-03-19 13:31:50 -05:00
base.py	off by one bateman	2025-03-18 08:40:43 -05:00
lora.py	naive model offloading support (handles automatically splitting parts of the model to requested device per memory constraints, either inferred or requested in the yaml, input tensors are automatically migrated to the right device, it SEEMS to work for training under the test trainer when split between GPU and CPU) (this was specifically only because that Flux imagegen model released so I can test it there)	2024-08-01 20:12:06 -05:00