Sentence Similarity
sentence-transformers
Safetensors
English
bert
ColBERT
multi-vector
feature-extraction
Generated from Trainer
dataset_size:497901
loss:Contrastive
text-embeddings-inference
Instructions to use NeuML/colbert-bert-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use NeuML/colbert-bert-tiny with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("NeuML/colbert-bert-tiny") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Inference
- Notebooks
- Google Colab
- Kaggle
| { | |
| "query_token_id": "[Q] ", | |
| "doc_token_id": "[D] ", | |
| "query_token": "[Q] ", | |
| "doc_token": "[D] ", | |
| "nprobe": 2, | |
| "ncandidates": 8192, | |
| "index_path": null, | |
| "nbits": 1, | |
| "kmeans_niters": 20, | |
| "similarity": "cosine", | |
| "bsize": 8, | |
| "accumsteps": 1, | |
| "lr": null, | |
| "maxsteps": null, | |
| "save_every": null, | |
| "resume": false, | |
| "warmup": null, | |
| "warmup_bert": null, | |
| "relu": false, | |
| "nway": 64, | |
| "use_ib_negatives": true, | |
| "query_maxlen": 32, | |
| "attend_to_mask_tokens": false, | |
| "dim": 128, | |
| "doc_maxlen": 180, | |
| "mask_punctuation": true, | |
| "checkpoint": null, | |
| "triples": null, | |
| "collection": null, | |
| "queries": null, | |
| "index_name": null, | |
| "overwrite": false, | |
| "root": "", | |
| "experiment": "default", | |
| "index_root": null, | |
| "name": "colbert-bert-tiny", | |
| "rank": 0, | |
| "nranks": 1, | |
| "amp": false, | |
| "gpus": 1 | |
| } | |