metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:10635
  - loss:MultipleNegativesSymmetricRankingLoss
base_model: sentence-transformers/multi-qa-mpnet-base-dot-v1
widget:
  - source_sentence: '12 Rules For Life: An Antidote to Chaos by Jordan B. Peterson'
    sentences:
      - Books on Investing
      - Books on Resilience
      - Books on Motivational
  - source_sentence: >-
      Get the Guy: Learn Secrets of the Male Mind to Find the Man You Want and
      the Love You Deserve by Matthew Hussey
    sentences:
      - Books on Complexity
      - Books on Decision Making
      - Books on Self-Help for Women
  - source_sentence: >-
      The Magic of Tiny Business (You Don’t Have to Go Big to Make a Great
      Living) by Sharon Rowe
    sentences:
      - Books on Vegetarianism
      - Books on Personal Development
      - Books on Emotions
  - source_sentence: >-
      The Dorito Effect: The Surprising New Truth About Food and Flavor by Mark
      Schatzker
    sentences:
      - Books on Skincare
      - Books on Work-Life Balance
      - Books on Problem Solving
  - source_sentence: '12 Rules For Life: An Antidote to Chaos by Jordan B. Peterson'
    sentences:
      - Books on Psychology
      - Books on Positive Thinking
      - Books on Investing
pipeline_tag: sentence-similarity
library_name: sentence-transformers

SentenceTransformer based on sentence-transformers/multi-qa-mpnet-base-dot-v1

This is a sentence-transformers model finetuned from sentence-transformers/multi-qa-mpnet-base-dot-v1 on the train dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: sentence-transformers/multi-qa-mpnet-base-dot-v1
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Dot Product
Training Dataset:
- train

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: MPNetModel 
  (1): Pooling({'word_embedding_dimension': 768, '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})
)

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 = [
    '12 Rules For Life: An Antidote to Chaos by Jordan B. Peterson',
    'Books on Psychology',
    'Books on Positive Thinking',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

train

Dataset: train
Size: 10,635 training samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 11 tokens
mean: 24.11 tokens
max: 60 tokens

min: 5 tokens
mean: 5.89 tokens
max: 10 tokens

	anchor	positive
type	string	string
details	min: 11 tokens mean: 24.11 tokens max: 60 tokens	min: 5 tokens mean: 5.89 tokens max: 10 tokens

Samples:

anchor	positive
`The Life-Changing Magic of Tidying Up: The Japanese Art of Decluttering and Organizing by Marie Kondō`	`Books on Organization`
`The Life-Changing Magic of Tidying Up: The Japanese Art of Decluttering and Organizing by Marie Kondō`	`Books on Minimalism`
`The Life-Changing Magic of Tidying Up: The Japanese Art of Decluttering and Organizing by Marie Kondō`	`Books on Japanese Art`

Loss: MultipleNegativesSymmetricRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Evaluation Dataset

train

Dataset: train
Size: 5,359 evaluation samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 8 tokens
mean: 22.0 tokens
max: 38 tokens

min: 4 tokens
mean: 5.85 tokens
max: 13 tokens

	anchor	positive
type	string	string
details	min: 8 tokens mean: 22.0 tokens max: 38 tokens	min: 4 tokens mean: 5.85 tokens max: 13 tokens

Samples:

anchor	positive
`12 Rules For Life: An Antidote to Chaos by Jordan B. Peterson`	`Books on Psychology`
`12 Rules For Life: An Antidote to Chaos by Jordan B. Peterson`	`Books on Self-Help`
`12 Rules For Life: An Antidote to Chaos by Jordan B. Peterson`	`Books on Personal Development`

Loss: MultipleNegativesSymmetricRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 2e-05
num_train_epochs: 10
warmup_ratio: 0.1

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 2e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 10
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	train loss
0.6006	200	2.6385	2.4890
1.2012	400	2.3324	2.4199
1.8018	600	2.1772	2.3891
2.4024	800	2.0635	2.3691
3.0030	1000	1.9915	2.3609
3.6036	1200	1.9008	2.3689
4.2042	1400	1.8603	2.3850
4.8048	1600	1.8421	2.3468
5.4054	1800	1.785	2.3649
6.0060	2000	1.786	2.3783
6.6066	2200	1.7331	2.3782
7.2072	2400	1.7062	2.3826
7.8078	2600	1.6929	2.3926
8.4084	2800	1.6618	2.4069
9.0090	3000	1.6348	2.4155
9.6096	3200	1.6553	2.4060

Framework Versions

Python: 3.10.12
Sentence Transformers: 4.1.0
Transformers: 4.52.4
PyTorch: 2.5.1+cu124
Accelerate: 1.8.1
Datasets: 3.6.0
Tokenizers: 0.21.2

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}