changAI-nomic-embed-text-finetuned

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: hyrinmansoor/changAI-nomic-embed-text-v1.5-finetuned
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Supported Modality: Text

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'ModernBertModel'})
  (1): Pooling({'embedding_dimension': 768, 'pooling_mode': 'mean', 'include_prompt': True})
  (2): Normalize({})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    'revenue from warenty claim items vs non-claimed lst yr',
    '[FIELD] item_code | [TABLE] tabWarranty Claim | desc: links to the Item master; identifies the product subject to a warranty claim',
    'Support_027',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000,  0.6816, -0.0659],
#         [ 0.6816,  1.0000,  0.0028],
#         [-0.0659,  0.0028,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 262,905 training samples
• Columns: anchor, positive, qid, and module
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	qid	module
  type	string	string	string	string
  details	min: 7 tokensmean: 16.09 tokensmax: 45 tokens	min: 6 tokensmean: 28.94 tokensmax: 71 tokens	min: 5 tokensmean: 6.44 tokensmax: 8 tokens	min: 3 tokensmean: 3.97 tokensmax: 5 tokens

• Samples:

  anchor	positive	qid	module
  how many days has the work order WO-999 been open?	[FIELD] name | [TABLE] tabWork Order | desc: Unique identifier for the production order.	Manufacturing_931	Manufacturing
  Whats the total amount invoiced for product XYZ?	[FIELD] item_code | [TABLE] tabInvoice Item | desc: Code of the item on the invoice line.	Selling_2982	Selling
  Which quality meetings are missing their corrective actions?	[FIELD] quality_meeting | [TABLE] tabQuality Action | desc: Links a quality action item back to the specific quality meeting where it was generated; if NULL for a meeting, indicates no actions were planned for it.	Quality Management_233	Quality Management

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "mini_batch_size": 4,
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode":