How to use GeneZC/bert-large-stsb with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, BertCls tokenizer = AutoTokenizer.from_pretrained("GeneZC/bert-large-stsb") model = BertCls.from_pretrained("GeneZC/bert-large-stsb", device_map="auto")
bert-large-uncased finetuned on STS-b.
bert-large-uncased
STS-b
batch size is 16, learning rate is 3e-5.
pearson_corr: 0.9024, spearman_corr: 0.8986
