Commit History
Correct typos in datasets.py (#639)
d1236f2
unverified
skip some flash attn patches unless explicitly enabled (#643)
895f0a0
unverified
use fastchat conversations template (#578)
e7d3e2d
unverified
update for recent transformers updates (#636)
60c7c48
unverified
attention_mask not needed for training (#642)
e8cbf50
unverified
eval_table isn't quite stable enough to be in default llama configs (#637)
d887ad8
unverified
Feat: Add support for upstream FA2 (#626)
19a600a
unverified
Added quotes to the pip install -e command to fix an incompatibility with shells that do glob expansion like zsh (#632)
5e5296a
unverified
Fernando Tarin Morales
commited on
Merge pull request #629 from OpenAccess-AI-Collective/chore/-change-default-model
f3d9390
unverified
Fix: Fail bf16 check when running on cpu during merge (#631)
cfbce02
unverified
default model changed
4fecbfe
Feat(doc): Add eval_sample_packing to doc (#625)
67b9888
unverified
tweak: improve base builder for smaller layers (#500)
923eb91
unverified
Maxime
commited on
better handling and logging of empty sharegpt turns (#603)
a363604
unverified
create a model card with axolotl badge (#624)
501958b
unverified
update README w deepspeed info (#605)
c25ba79
unverified
chore(callback): Remove old peft saving code (#510)
d5f8589
unverified
misc fixes to add gptq tests (#621)
03e5907
unverified
split completion text to sequence_len (#616)
97d3776
unverified
run eval on the first step to get a baseline (#617)
2844eb2
unverified
let MAX_JOBS use the default since we're not resource constrained on our self-hosted runners (#427)
e85d2eb
unverified
skip the gpu memory checks if the device is set to 'auto' (#609)
196ff11
unverified
ignore wandb to resolve isort headaches (#619)
92512c3
unverified
fix distributed devices (#612)
2fe95cd
unverified
Maxime
commited on
Create multi-node.md (#613)
c1382e7
unverified
Maxime
commited on
Only run tests when a change to python files is made (#614)
5d931cc
unverified
Update requirements.txt (#610)
ec0958f
unverified
Javier
commited on
support to disable exllama for gptq (#604)
faecff9
unverified
Delete duplicate lines (#606)
aa656e0
unverified
update dockerfile to not build evoformer since it fails the build (#607)
b53e777
unverified
more sane defaults for openllama 3b used for quickstarts (#602)
674c576
unverified
improve handling for empty text on the tokenization step (#502)
1eebbd0
unverified
Fix for check with cfg and merge_lora (#600)
62a7741
unverified
minor tweaks to simplify (#597)
31b9e0c
unverified
btlm and falcon monkey patches for flash attn (#566)
6b9b229
unverified
add bf16 check (#587)
131afdb
unverified
Feat(data): Allow loading local csv and text (#594)
00dce35
unverified
gather/broadcast the max value of the packing efficiency automatically (#463)
b15b19e
unverified
don't add position_ids for evals (#591)
ab534d7
unverified
optionally configure sample packing for evals (#589)
21ec195
unverified
make phi training work with Loras (#588)
62eaee7
unverified
set fsdp state dict (#584)
be75668
unverified
Jan Philipp Harries
Jan Philipp Harries
commited on
pop block_cls since it's not an actual kwarg
aeec7c4
don't resize embeddings if it's already large enough (#577)
3607882
unverified
Support Sample packing for phi arch (#586)
12a2dbb
unverified
Feat(doc): Add features to doc (#583)
3a2edc8
unverified
support custom field for completion from yml (#580)
f7a2263
unverified
Fix Codellama examples (#582)
1aa4007
unverified
Doan Minh Phuong
commited on