Update qlora.yml - remove `max_packed_sequence_len` (#1210) [skip ci] 5407ddd unverified 7flash commited on Jan 26
drop py39 docker images, add py311, upgrade pytorch to 2.1.2 (#1205) 74c72ca unverified winglian commited on Jan 26
more checks and fixes for deepspeed and fsdp (#1208) [skip ci] e923e62 unverified winglian commited on Jan 26
workaround for transformers bug requireing do_sample for saveing pretrained (#1206) ba944e6 unverified winglian commited on Jan 25
make sure to register the base chatml template even if no system message is provided (#1207) badda37 unverified winglian commited on Jan 25
precompute dpo logprobs setting and fixes (#1199) [skip ci] 33e1170 unverified winglian commited on Jan 25
fix learning rate scheduler's warnings (#1135) [skip ci] b4ac96a unverified ricdomolm winglian commited on Jan 25
Feat/chatml add system message (#1117) 98b4762 unverified mhenrichsen Mads Henrichsen winglian commited on Jan 25
fix(log): improve warning to clarify that lora_modules_to_save expect a list (#1197) 08719b9 unverified Nanobit commited on Jan 25
Standardize system prompt format for AlpacaPrompter (#1190) [skip ci] af02430 unverified Oleh Kuznetsov commited on Jan 24
upgrade deepspeed to 0.13.1 for mixtral fixes (#1189) [skip ci] 8a49309 unverified winglian commited on Jan 24
more dpo fixes for dataset loading and docs (#1185) [skip ci] 5bce45f unverified winglian commited on Jan 24
Fix generation_config validation raises Exception for do_merge_lora (#1184) 02f2c72 unverified tisorlawan commited on Jan 24
Add support for offline mode with HF_HUB_OFFLINE envvar (#1182) 71141de unverified James Wade winglian commited on Jan 24
don't fail if can't cast weights due to offload when merging (#1172) [skip ci] fb7f9b9 unverified winglian commited on Jan 23
Fine-Tuning Mistral-7b for Real-World Chatbot Applications Using Axolotl (Lora used) (#1155) cc25039 unverified Tilemachos Chatzipapas twenty8th winglian commited on Jan 23
Feat(test): Add tests for alpaca chatml prompt tokenizer (#1088) 5439707 unverified JohanWork Nanobit commited on Jan 23
support for explicit test_dataset definition for evals (#786) cda52dc unverified winglian commited on Jan 23
add commit message option to skip docker image builds in ci (#1168) [skip ci] 0f77b8d unverified winglian commited on Jan 23
improve vram use w gradient checkpointing (#1167) [skip ci] 802f966 unverified winglian commited on Jan 23
Add mlflow callback for pushing config to mlflow artifacts (#1125) b8e5603 unverified JohanWork commited on Jan 22
set fp16 to false if bf16, update bf16: auto in example YAMLs (#1122) [skip ci] 782b6a4 unverified winglian Nanobit commited on Jan 22
make sure the model config loader respects the model_revision too (#1160) [skip-ci] fccb542 unverified winglian commited on Jan 22
feat(dataset): add config to keep processed dataset in memory (#1152) 3db5f2f unverified Nanobit commited on Jan 20
Add shifted sparse attention (#973) [skip-ci] 1d70f24 unverified jrc joecummings winglian commited on Jan 18
fix(preprocess): Make sure dataset not loaded from cache when using preprocess cli (#1136) 1e56b88 unverified Nanobit commited on Jan 17