VictorSanh MoritzLaurer HF staff commited on
Commit
4b1a6e0
1 Parent(s): 4ceecdb

fix small typo (#2)

Browse files

- fix small typo (c14f48116ba8c527e05d70216058166175ae9e38)


Co-authored-by: Moritz Laurer <MoritzLaurer@users.noreply.huggingface.co>

Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -108,7 +108,7 @@ Following this, we perform instruction fine-tuning on [The Cauldron](https://hug
108
  - [atlas-math-sets](https://huggingface.co/datasets/AtlasUnified/atlas-math-sets)
109
  - [goat](https://huggingface.co/datasets/tiedong/goat)
110
 
111
- We use Lora to train the parameters initialized from pre-trained backbones and full fine-tuning for newly initialized parameters (modality connector), as we find this strategy to be more stable as long as more computationally efficient.
112
 
113
  More details (training procedure, data selection, hyper-parameters, etc.) along with lessons learned from our ablations will be available in an upcoming technical report.
114
 
 
108
  - [atlas-math-sets](https://huggingface.co/datasets/AtlasUnified/atlas-math-sets)
109
  - [goat](https://huggingface.co/datasets/tiedong/goat)
110
 
111
+ We use Lora to train the parameters initialized from pre-trained backbones and full fine-tuning for newly initialized parameters (modality connector), as we find this strategy to be more stable as well as more computationally efficient.
112
 
113
  More details (training procedure, data selection, hyper-parameters, etc.) along with lessons learned from our ablations will be available in an upcoming technical report.
114