gte-large-en-v1.5 finetuned for TeleQnA response scoring

base model: Alibaba-NLP/gte-large-en-v1.5

We finetuned the model to help compute (approximate) soft scores for textual entailment to help find the option that entails the response from a model whose response is not conditioned on the options.

Usage (Sentence-Transformers)

This is a sentence-transformers model: It maps sentences & paragraphs to a 1024 dimensional dense vector space and can be used for tasks like clustering or semantic search. Using this model becomes easy when you have sentence-transformers installed:

pip install -U sentence-transformers

Then you can use the model like this:

from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]

model = SentenceTransformer('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)

Loss:

sentence_transformers.losses.TripletLoss.TripletLoss with parameters:

{'distance_metric': 'TripletDistanceMetric.EUCLIDEAN', 'triplet_margin': 5}

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: NewModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Citing & Authors

We simply finetune the GTE large v1.5 model found at: Alibaba-NLP/gte-large-en-v1 Link to the paper: https://huggingface.co/papers/2308.03281