Support Sample packing for phi arch (#586)

Fix Codellama examples (#582)

Phi examples (#569)

Add training callback to send predictions to WandB table (#521)

recommend padding when using sample packing (#531)

Add support for GPTQ using native transformers/peft (#468)

pad_to_worst_case_seq_len boolean, for testing memory limits (#498)

Feat(cfg): Add code-llama configs for all sizes (#479)

Add example Llama 2 ReLoRA config (#471)

improve llama pad token handling (#475)

don't use mask expansion for inference (#392)

new llama-2 default settings (#370)

Add wandb_entity to wandb options, update example configs, update README (#361)

set group_by_length to false in examples

feat/llama-2 examples (#319)

Add XGen info to README and example config

Use AutoTokenizer for redpajama example

Merge pull request #92 from OpenAccess-AI-Collective/flash-optimum

Merge branch 'main' into flash-optimum

tweak config to work

Merge pull request #193 from OpenAccess-AI-Collective/config-fixes-20230612

Fix config path after config moved

config fixes

forgot to add this file

update openllama and clean up paths

more pruning

more config pruning and migrating

get rid of some configs, formalize pythioa lora config

address PR feedback

linting fix

use pythia-12b, neox-20b is flaky

Merge pull request #132 from utensil/falcon-7b-qlora

Trim trailing whitespace

Default `wandb_project` to empty as suggested

Default micro_batch_size to 1 for a safer start

Add comments/alternatives for falcon-qlora configs

falcon + qlora + xformer mbs 40 gas 2 on A6000

Create config-7b-qlora.yml

swap batch size for gradient accumulation steps to decouple from num gpu

Merge pull request #105 from viktoriussuwandi/viktoriussuwandi-patch

Merge pull request #106 from fearnworks/qlora-openllama-3b-example

Lint and format

Update examples/qlora-openllama-3b/README.md

Add Readme, Clean up comments

Add qlora-openllama-3b example

Update wandb_log_model on config.yml

Update wandb_log_model on config-3b.yml

Update wandb_log_model on config.yml

Merge branch 'main' into refactor/rename-4b-to-gptq

sane starter defaults and add lora