SentenceTransformer based on intfloat/multilingual-e5-large-instruct

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: intfloat/multilingual-e5-large-instruct
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • measuring-embeddings-v3

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

Training Details

Training Dataset

measuring-embeddings-v3

• Dataset: measuring-embeddings-v3 at 1b3cbbe
• Size: 7,552 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 9 tokens mean: 15.96 tokens max: 40 tokens	min: 120 tokens mean: 255.56 tokens max: 512 tokens	min: 0.0 mean: 0.22 max: 0.95

• Samples:

  sentence1	sentence2	score
  How can I combine the sub-query with the main query to fetch the last uncertainty report?	What do measurement equipment measure?[object Object]Each equipment measures a physical magnitude, also known as a variable. Based on the type of variable they measure, devices are classified into different categories.[object Object][object Object]Equipment classification:[object Object]- Primary meter: Assigned by default to equipments like orifice plates.[object Object]- Secondary meter: Assigned by default to equipments like transmitters.[object Object]- Tertiary meter: Used for other types of equipments.[object Object][object Object]Equipment types in the database:[object Object]The database includes a table listing all equipment types. Examples of equipment types are:[object Object]- Differential pressure transmitters[object Object]- RTDs (Resistance Temperature Detectors)[object Object]- Orifice plates[object Object]- Multivariable transmitters[object Object]- Ultrasonic meters[object Object][object Object]Meteorological checks for equipments:[object Object]Each equipment type is assigned a meteorological check, which can be either:[object Object]- Calibration: To ensure measurement accuracy.[object Object]- Inspection: To verify proper functioning.[object Object][object Object]Data storage in tables:[object Object]The database also includes a separate table for equipment classific...	0.1
  What is the column name for the calibration date in the calibration table?	How are flow computers and measurement systems related?[object Object]Flow computers can have multiple systems assigned to them. However, a measurement system can only be assigned to one flow computer.[object Object][object Object]Database terminology:[object Object]In the database, this relationship is referred to as:[object Object]- Meter streams[object Object]- Meter runs[object Object]- Sections[object Object][object Object]Storage of the relationship:[object Object]The relationship between a flow computer and its assigned measurement system is stored in a special table.[object Object][object Object]User context:[object Object]When a user refers to a "meter stream," they are indicating that they are searching for a measurement system assigned to a specific flow computer.	0.1
  What is the name of the table that contains the flow computer tags?	What is equipment calibration?[object Object]Calibration is a metrological verification process used to ensure the accuracy of measurement equipment. It is performed periodically, based on intervals set by the company or a regulatory body.[object Object][object Object]Purpose of calibration:[object Object]The calibration process corrects any deviations in how the equipment measures physical magnitudes (variables). This ensures the equipment provides accurate and reliable data.[object Object][object Object]Calibration cycles:[object Object]There are two main calibration cycles:[object Object]1. As-found: Represents the equipment's measurement accuracy before any adjustments are made. This cycle is almost always implemented.[object Object]2. As-left: Represents the equipment's measurement accuracy after adjustments are made. This cycle is used depending on regulatory requirements.[object Object][object Object]Calibration uncertainty:[object Object]- Uncertainty is included in the results of a calibration.[object Object]- Calibration uncertainty refers to the margin of error in the device's measurements, which also affects the uncertainty of the measured variable or ...	0.05

• Loss: CoSENTLoss with these parameters:
  json

  1{ 2 "scale": 20.0, 3 "similarity_fct": "pairwise_cos_sim" 4}

Evaluation Dataset

measuring-embeddings-v3

• Dataset: measuring-embeddings-v3 at 1b3cbbe
• Size: 1,618 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 9 tokens mean: 15.83 tokens max: 40 tokens	min: 120 tokens mean: 250.41 tokens max: 512 tokens	min: 0.0 mean: 0.23 max: 0.95

• Samples:

  sentence1	sentence2	score
  Identify any additional tables or columns that might be needed for the query.	How are flow computers and measurement systems related?[object Object]Flow computers can have multiple systems assigned to them. However, a measurement system can only be assigned to one flow computer.[object Object][object Object]Database terminology:[object Object]In the database, this relationship is referred to as:[object Object]- Meter streams[object Object]- Meter runs[object Object]- Sections[object Object][object Object]Storage of the relationship:[object Object]The relationship between a flow computer and its assigned measurement system is stored in a special table.[object Object][object Object]User context:[object Object]When a user refers to a "meter stream," they are indicating that they are searching for a measurement system assigned to a specific flow computer.	0.2
  What columns in these tables contain the measurement system tag and the flow computer tag?	How does a flow computer generate and store reports?[object Object]A flow computer generates daily or hourly reports to provide users with operational data. These reports are stored in the flow computer's memory in an organized format.[object Object][object Object]Report structure:[object Object]- Each report includes:[object Object]- Date and time of the data recording.[object Object]- Data recorded from flow computers.[object Object][object Object]Data storage in tables:[object Object]The reports are saved in two tables:[object Object]1. Main table (Index):[object Object] - Stores the date, time, and flow computer identifier.[object Object]2. Detail table:[object Object] - Stores the measured values associated with the report.[object Object][object Object]Connection to the Modbus table:[object Object]The flow computer's reports are linked to a Modbus table. This table contains the names corresponding to each value in the reports, making it easier to interpret the data.	0.1
  Identify the column that stores the calibration number.	What kind of data store an equipment?[object Object]Equipments can capture meteorological data, such as pressure, temperature, and volume (magnitudes). This data is essential for users to perform various calculations.[object Object][object Object]Data storage:[object Object]- The measured values are stored in a special table in the database for magnitudes. This table contains the values of the variables captured by the equipments.[object Object]- These values are [object Object] from the fluid (e.g., raw pressure, temperature, or volume readings). [object Object], such as uncertainty.[object Object]- The values stored in the variable values table are [object Object] from variable uncertainty values, which are calculated separately and represent the margin of error.[object Object][object Object]Accessing the data:[object Object]- Users typically access the data by referring to the readings from the measurement system, not directly from the individual equipments.[object Object]- The readings are stored in a "variable values" table within the database.[object Object][object Object]Linking variable names:[object Object]If the user needs to kno...	0.1

• Loss: CoSENTLoss with these parameters:
  json

  1{ 2 "scale": 20.0, 3 "similarity_fct": "pairwise_cos_sim" 4}

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 7
• per_device_eval_batch_size: 7
• gradient_accumulation_steps: 4
• learning_rate: 3e-05
• num_train_epochs: 20
• warmup_ratio: 0.1

All Hyperparameters

Training Logs

Framework Versions

• Python: 3.11.0
• Sentence Transformers: 3.4.0
• Transformers: 4.48.1
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.2.0
• Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers

CoSENTLoss