set fp16 to false if bf16, update bf16: auto in example YAMLs (#1122) [skip ci] 782b6a4 unverified winglian Nanobit commited on Jan 22
Add shifted sparse attention (#973) [skip-ci] 1d70f24 unverified jrc joecummings winglian commited on Jan 18
new evals_per_epoch and saves_per_epoch to make things cleaner (#944) 5f79b82 unverified winglian commited on Dec 12, 2023
don't compile deepspeed or bitsandbytes from source (#837) f544ab2 unverified winglian commited on Nov 9, 2023
simplify by removing duplicate base_model_config (#772) 2d8def6 unverified winglian commited on Oct 23, 2023
prepared dataset caching, other misc fixes (#665) e50a64e unverified winglian commited on Oct 3, 2023
eval_table isn't quite stable enough to be in default llama configs (#637) d887ad8 unverified winglian commited on Sep 26, 2023
Add training callback to send predictions to WandB table (#521) 5b67ea9 unverified Glavin001 commited on Sep 13, 2023
recommend padding when using sample packing (#531) 3437149 unverified winglian commited on Sep 6, 2023
Add support for GPTQ using native transformers/peft (#468) 3355706 unverified winglian commited on Sep 5, 2023
new llama-2 default settings (#370) fdffef5 unverified mhenrichsen Mads Henrichsen commited on Aug 14, 2023
Add wandb_entity to wandb options, update example configs, update README (#361) 7019509 unverified Morgan McGuire Morgan McGuire winglian commited on Aug 12, 2023