 Matthias Reso
|
113ea18bf1
Replace LlamaTokenizer with AutoTokenizer
|
11 months atrás |
 Hamid Shojanazeri
|
11f51db28c
adding the kbit prep in the code
|
11 months atrás |
|
Hamid Shojanazeri
|
f058ff6ccd
update due to peft new release
|
11 months atrás |
|
Hamid Shojanazeri
|
ffdc93f00a
Merge branch 'main' into wandb_logging
|
1 year atrás |
|
Matthias Reso
|
c5a382e509
Make tests run on cpu only machines
|
1 year atrás |
|
Hamid Shojanazeri
|
43ea6bfa71
Merge branch 'main' into ssdp
|
1 year atrás |
|
Hamid Shojanazeri
|
d51d2cce9c
adding sdpa for flash attn
|
1 year atrás |
|
Hamid Shojanazeri
|
9a2434f408
update the fine-tuning script
|
1 year atrás |
|
Hamid Shojanazeri
|
162be4c045
Revert "Flop counter, profiling and GC (#357)"
|
1 year atrás |
|
Hamid Shojanazeri
|
71d137c722
Merge branch 'main' into flop_counter_gc
|
1 year atrás |
 Less Wright
|
3f2c33e4f8
Update finetuning.py - remove nightly check
|
1 year atrás |
|
Hamid Shojanazeri
|
b15ffeeaf4
clean up
|
1 year atrás |
|
Hamid Shojanazeri
|
19089269d3
add gc
|
1 year atrás |
 kldarek
|
989b6ee812
wandb logging feedback
|
1 year atrás |
 Abhilash Majumder
|
4793f0fdf3
Merge branch 'main' into ipex_feature
|
1 year atrás |
|
kldarek
|
fc5485d916
fixing wandb for fsdp
|
1 year atrás |
|
kldarek
|
f2406cac07
cleanup spaces
|
1 year atrás |
|
kldarek
|
cf373529f7
basic wandb logging instrumentation
|
1 year atrás |
|
Hamid Shojanazeri
|
aa24e8f57e
fix typos
|
1 year atrás |
|
Hamid Shojanazeri
|
38c8cf08c4
adding hsdp
|
1 year atrás |
|
Abhilash Majumder
|
6a78b96764
Merge branch 'main' into ipex_feature
|
1 year atrás |
|
Matthias Reso
|
4c225c65eb
Fix order of concat vs sampler
|
1 year atrás |
|
Matthias Reso
|
5da84b2913
Fix usage of dataclass for train_config and fsdp_config
|
1 year atrás |
|
Matthias Reso
|
653a79e3dd
Invalidate context in labels for samsum + grammar
|
1 year atrás |
|
Matthias Reso
|
a647955fc8
Make packing/padding a training setting
|
1 year atrás |
|
Matthias Reso
|
2e4bd2a665
Resize vocab size to fix idx error
|
1 year atrás |
|
Matthias Reso
|
ca41c1c697
Adjust tests to len based batch sampling
|
1 year atrás |
 Shijie Wu
|
91e2573aa8
pass weight_decay into optimizer
|
1 year atrás |
 Howard Liberty
|
cc356b6017
Add FSDP CPU offloading option
|
1 year atrás |
 abhilash1910
|
ad6b27d316
merge conflicts
|
1 year atrás |
