Hugging Face's logo

ZurichNLP
/
unsup-simcse-xlm-roberta-base

Feature Extraction
Transformers
PyTorch
Safetensors
xlm-roberta
Model card Files
xet
 Community
1
jvamvas commited on
Commit
91d8edf
·
1 Parent(s): 8f206e1

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -1
README.md CHANGED
@@ -102,7 +102,10 @@ pipeline_tag: feature-extraction
102
 
103
  See a similar English model released by Gao et al.: https://huggingface.co/princeton-nlp/unsup-simcse-roberta-base
104
 
105
- Fine-tuning was done using the [reference implementation of SimCSE](https://github.com/princeton-nlp/SimCSE) and the 1M sentences from English Wikipedia released by the authors:
 
 
 
106
  ```bash
107
  python train.py \
108
  --model_name_or_path xlm-roberta-base \
 
102
 
103
  See a similar English model released by Gao et al.: https://huggingface.co/princeton-nlp/unsup-simcse-roberta-base
104
 
105
+ Fine-tuning was done using the [reference implementation of SimCSE](https://github.com/princeton-nlp/SimCSE) and the 1M sentences from English Wikipedia released by the authors.
106
+ As a sentence representation we use the average of the last hidden states (`pooler_type=avg`), which ensures compatibility to Sentence-BERT.
107
+
108
+ Fine-tuning command:
109
  ```bash
110
  python train.py \
111
  --model_name_or_path xlm-roberta-base \