issues
search
jina-ai
/
jerboa
LLM finetuning
Apache License 2.0
41
stars
4
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
We need to train alpaca-lora on the same number of lora layers to be able to compare it to falcon 7B and understand the effect of changing from llama to falcon
#71
alaeddine-13
closed
1 year ago
0
Compare code aligned model to current SOTA
#70
azayz
closed
1 year ago
0
chore: bump transformers and all
#69
samsja
closed
1 year ago
0
Create an 8bits model for inference out of falcon 40b
#68
samsja
closed
1 year ago
0
feat: script to execute training run and stop runpod afterwards
#67
sebastian-weisshaar
closed
1 year ago
0
fix: readme cli after switching to typer
#66
samsja
closed
1 year ago
0
Chore update dependency
#65
samsja
closed
1 year ago
0
Add lima dataset to the training pipeline
#64
alaeddine-13
closed
1 year ago
1
feat: update peft ot latest main
#63
samsja
closed
1 year ago
0
chore: random name with model and data path name
#62
alaeddine-13
closed
1 year ago
1
feat: add type hint to load model
#61
samsja
closed
1 year ago
0
docs: add runpod setup
#60
sebastian-weisshaar
closed
1 year ago
0
Fix bug in save_pretrained
#59
samsja
closed
1 year ago
0
feat: allow customizing dataset and dataset processing and support redpajamas instruct datasets
#58
alaeddine-13
closed
1 year ago
0
refactor: remove useless file
#57
samsja
closed
1 year ago
0
feat: add typer and logging of config
#56
samsja
closed
1 year ago
0
feat: docker image
#55
sebastian-weisshaar
closed
1 year ago
0
feat: docker container for runpod
#54
sebastian-weisshaar
closed
1 year ago
0
Add dockerfile to jerboa to run on runpod
#53
samsja
closed
1 year ago
0
logs dataset in wandb
#52
samsja
closed
1 year ago
0
Pipeline training dataset refactoring
#51
alaeddine-13
closed
1 year ago
0
refactor: allow to work with any causal LLM including falcon
#50
sebastian-weisshaar
closed
1 year ago
0
Add red pajamas instruct dataset to our pipeline
#49
alaeddine-13
closed
1 year ago
0
add redpajama 7b to our pipeline
#48
samsja
closed
1 year ago
1
chore: add eval libraries
#47
azayz
closed
1 year ago
0
feat: add code eval
#46
azayz
closed
1 year ago
0
Incorporate baize data and training into jerboa training pipeline
#45
alaeddine-13
closed
1 year ago
0
chore: add eval for code
#44
azayz
closed
1 year ago
0
feat: add more layers to lora finetuning
#43
alaeddine-13
closed
1 year ago
1
fix: remove checkpointing
#42
samsja
closed
1 year ago
0
feat: code alpaca dataset
#41
alaeddine-13
closed
1 year ago
0
fix: changed names in evaluate and in test
#40
sebastian-weisshaar
closed
1 year ago
0
fix: do not raise an exception if 8bit on cpu
#39
sebastian-weisshaar
closed
1 year ago
0
Prepare our code base to be able to ft on code alpaca
#38
samsja
closed
1 year ago
0
refactor: change default training to 8bit
#37
sebastian-weisshaar
closed
1 year ago
0
Publish alpaca lora 8 bits on our HF account
#36
samsja
closed
1 year ago
1
Long term: create a good evaluation for QA code
#35
samsja
closed
1 year ago
8
Align falcon 7b on alpaca
#34
samsja
closed
1 year ago
0
Align llama7 on code alpaca
#33
samsja
closed
1 year ago
1
Align falcon 40b on code alpaca
#32
samsja
closed
1 year ago
1
Allign llama 7b on alpaca with 4 bits
#31
samsja
closed
1 year ago
0
refactor: merge eval files
#30
sebastian-weisshaar
closed
1 year ago
1
refactor: combine two eval files into one, delete original ones
#29
sebastian-weisshaar
closed
1 year ago
1
fix: run eval during test
#28
samsja
closed
1 year ago
0
fix: evaluation
#27
azayz
closed
1 year ago
0
fix: add eval alpaca
#26
azayz
closed
1 year ago
0
feat: 4bit implementation for training
#25
sebastian-weisshaar
closed
1 year ago
1
Feat q lora implementation
#24
sebastian-weisshaar
closed
1 year ago
0
fix: switch to yahma
#23
azayz
closed
1 year ago
0
Add qlora to our current codebase
#22
samsja
closed
1 year ago
1
Previous
Next