README.md

---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:600
- loss:CoSENTLoss
- dataset_size:2500
base_model: Intellexus/mbert-tibetan-continual-wylie-final
widget:
- source_sentence: gong bu gzhan min de de'i min//
  sentences:
  - de la byang chub kyi yan lag bdun gang zhe na/ 'di lta ste/ dran pa yang dag byang
    chub kyi yan lag dang / chos rab rnam 'byed yang dag byang chub kyi yan lag dang
    / brtson 'grus yang dag byang chub kyi yan lag dang / dga' ba yang dag byang chub
    kyi yan lag dang / shin tu sbyangs pa yang dag byang chub kyi yan lag dang / ting
    nge 'dzin yang dag byang chub kyi yan lag dang / btang snyoms yang dag byang chub
    kyi yan lag ste/ de dag ni byang chub kyi yan lag bdun ces bya'o//
  - phung myin gal te de de myin//
  - sha ra dwa ti'i bu gzhan yang byang chub sems dpa' sems dpa' chen po byang sa
    las 'da' bar 'dod pas/ shes rab kyi pha rol tu phyin pa la bslab par bya'o//
- source_sentence: kun rdzob tu ni thugs brtse bas// rgyu mthun de dag thub pa bzhed//
  sentences:
  - yang na sku gzugs ma nyams spyan ras sngar zlas
  - kun rdzob 'jig rten grags pa la// brtan na tshad ma'i rnam gzhag 'gal//
  - "gzhan gyi dbang gi ngo bo nyid//\r\nrnam rtog yin te rkyen las byung //\r\ngrub\
    \ ni de la snga ma po//\r\nrtag tu med par gyur pa gang //"
- source_sentence: 'bdag las ma yin gzhan las min// gnyis las ma yin rgyu med min//
    dngos po gang dag gang na yang // skye ba nam yang yod ma yin//

    '
  sentences:
  - 'shing rta che bu sems can che//


    rtag mo bkres mthong stag phrug rnams//


    thar bar bya phyir snying rje yis//'
  - 'phyogs chos de chas khyab pa yi// gtan tshigs de ni rnam gsum nyid// med na mi
    ''byung nges phyir ro// gtan tshigs ltar snang de las gzhan//

    '
  - sems can rnams kyi 'dod chags byang gyur cig//
- source_sentence: gang gi tshe rgyal po pad ma chen po dpung dang mthu che ba de'i
    tshe na/ des kyang dpung gi tshogs yan lag bzhi pa/ glang po che pa'i tshogs dang
    / rta pa'i tshogs dang / shing rta pa'i tshogs dang / dpung bu chung gi tshogs
    go bskon te/ yul ang ga tsam pa ma gtogs pa bcom nas phyir ldog par byed do//
  sentences:
  - de tshe rig pa'i rgyal po bsgrub// gal te de ni rab byung gyur// sdom pa gsum
    la yang dag gnas// so sor thar dang byang chub sems// rig 'dzin sdom pa mchog
    yin no//
  - spyir theg pa zhes bya ba'i nges tshig ni/ ya na zhes bya ba 'gro ba'i bya ba
    ston pa'i tshig yin pas tshig gzugs por lam la bya'o//
  - rgyal po chen po 'di ltar yang dge sbyong dang / bram ze kha cig dad pas byin
    pa dag spyad nas ltad mo sna tshogs rtsom pa la sbyor bar brtson pas gnas pa 'di
    lta ste/
- source_sentence: dam tshig nyams pa'i nyes pa ni/ 'dod pa'i phyogs mi 'grub cing
    / mi 'dod pa'i phyogs rnams thob pa ste/
  sentences:
  - 'dam tshig dang ni mi ldan na// bsgrubs kyang ''grub par mi ''gyur te//

    rgyu med pa yi ''bras bu bzhin// tshe yi  dus byas dmyal bar ''gro//

    '
  - rang sangs rgyas rnams kyi rnam par grol ba ni/ ngag gi lam dang bral ba las skyes
    pa/
  - 'lha dang lha mo ji lta bas// bdud rtsi''i  bum pas dbang bskur ba//

    chu''i dgongs pa  ye shes lnga''i// rtags su  sku lnga rdzogs pa''o//

    '
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
model-index:
- name: SentenceTransformer based on Intellexus/mbert-tibetan-continual-wylie-final
  results:
  - task:
      type: semantic-similarity
      name: Semantic Similarity
    dataset:
      name: Unknown
      type: unknown
    metrics:
    - type: pearson_cosine
      value: 0.8350341193647188
      name: Pearson Cosine
    - type: spearman_cosine
      value: 0.8539838973084938
      name: Spearman Cosine
---

# SentenceTransformer based on Intellexus/mbert-tibetan-continual-wylie-final

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Intellexus/mbert-tibetan-continual-wylie-final](https://huggingface.co/Intellexus/mbert-tibetan-continual-wylie-final). 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:** [Intellexus/mbert-tibetan-continual-wylie-final](https://huggingface.co/Intellexus/mbert-tibetan-continual-wylie-final) <!-- at revision ed345c6d5cdee3f8ca31c40ff9940e56cb0c3f2d -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 768 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->

### Model Sources

- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)

### Full Model Architecture

```
SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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})
)
```

## Usage

### Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

```bash
pip install -U sentence-transformers
```

Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    "dam tshig nyams pa'i nyes pa ni/ 'dod pa'i phyogs mi 'grub cing / mi 'dod pa'i phyogs rnams thob pa ste/",
    "dam tshig dang ni mi ldan na// bsgrubs kyang 'grub par mi 'gyur te//\nrgyu med pa yi 'bras bu bzhin// tshe yi  dus byas dmyal bar 'gro//\n",
    "lha dang lha mo ji lta bas// bdud rtsi'i  bum pas dbang bskur ba//\nchu'i dgongs pa  ye shes lnga'i// rtags su  sku lnga rdzogs pa'o//\n",
]
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]
```

<!--
### Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details>
-->

<!--
### Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details>
-->

<!--
### Out-of-Scope Use

*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->

## Evaluation

### Metrics

#### Semantic Similarity

* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)

| Metric              | Value     |
|:--------------------|:----------|
| pearson_cosine      | 0.835     |
| **spearman_cosine** | **0.854** |

<!--
## Bias, Risks and Limitations

*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->

<!--
### Recommendations

*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->

## Training Details

### Training Dataset

#### Unnamed Dataset

* Size: 2,500 training samples
* Columns: <code>text1</code>, <code>text2</code>, and <code>label</code>
* Approximate statistics based on the first 1000 samples:
  |         | text1                                                                             | text2                                                                             | label                                                           |
  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------|
  | type    | string                                                                            | string                                                                            | float                                                           |
  | details | <ul><li>min: 6 tokens</li><li>mean: 19.74 tokens</li><li>max: 67 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 22.11 tokens</li><li>max: 83 tokens</li></ul> | <ul><li>min: 0.02</li><li>mean: 0.51</li><li>max: 1.0</li></ul> |
* Samples:
  | text1                                                 | text2                                                                                                 | label               |
  |:------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:--------------------|
  | <code>'on pa rnams kyang rna bas sgra thos p</code>   | <code>'on pa rnams rna bas sgra thes par bya'o snyam pa dang / smyon pa rnams dran pa thob par</code> | <code>0.5</code>    |
  | <code>com ldan 'das de bzhin gshegs pa dgra bc</code> | <code>mkhas pa yongs su gzung bar 'dod pa'i byang chub sems dpa' sems dpa' chen</code>                | <code>0.229</code>  |
  | <code>pa /sems can thams cad</code>                   | <code>ng / snying rje'i sems dang ldan pa</code>                                                      | <code>0.3335</code> |
* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
  ```json
  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  ```

### Evaluation Dataset

#### Unnamed Dataset

* Size: 150 evaluation samples
* Columns: <code>text1</code>, <code>text2</code>, and <code>label</code>
* Approximate statistics based on the first 150 samples:
  |         | text1                                                                              | text2                                                                              | label                                                         |
  |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:--------------------------------------------------------------|
  | type    | string                                                                             | string                                                                             | float                                                         |
  | details | <ul><li>min: 8 tokens</li><li>mean: 32.74 tokens</li><li>max: 126 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 32.12 tokens</li><li>max: 121 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.5</li><li>max: 1.0</li></ul> |
* Samples:
  | text1                                                                                                                  | text2                                                                           | label               |
  |:-----------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------|:--------------------|
  | <code>khang ljon shing rgyal mtshan seng ge rta</code>                                                                 | <code>khang bzangs ljong shing bram ze seng ge rta</code>                       | <code>0.5625</code> |
  | <code>rnam par thar pa'i sgo mtshan ma med pa/</code>                                                                  | <code>yod ces bya bar yang dag par rjes su mi mthong ba/</code>                 | <code>0.375</code>  |
  | <code>byang chub ni chos kyi dbyings kyi gnas kyis gnas pa'o// byang chub ni de bzhin nyid rjes su rtogs pa'o//</code> | <code>nges pa yod na mngon sum min// 'dra bar 'dzin pa rtog pa yin//<br></code> | <code>0.0</code>    |
* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
  ```json
  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  ```

### Training Hyperparameters
#### Non-Default Hyperparameters

- `eval_strategy`: epoch
- `per_device_train_batch_size`: 32
- `gradient_accumulation_steps`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 7
- `load_best_model_at_end`: True

#### All Hyperparameters
<details><summary>Click to expand</summary>

- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: epoch
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 8
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 16
- `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`: 7
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `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`: True
- `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}
- `tp_size`: 0
- `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
- `dispatch_batches`: None
- `split_batches`: 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

</details>

### Training Logs
| Epoch  | Step | Training Loss | Validation Loss | spearman_cosine |
|:------:|:----:|:-------------:|:---------------:|:---------------:|
| 1.0    | 2    | 56.9409       | 2.7480          | 0.8357          |
| 2.0    | 4    | 53.1489       | 2.7016          | 0.8412          |
| 3.0    | 6    | 52.3657       | 2.6812          | 0.8462          |
| 3.8421 | 7    | 89.1774       | 2.6767          | 0.8471          |
| 0.8101 | 4    | 96.7978       | 2.7350          | 0.8455          |
| 1.8101 | 8    | 94.8279       | 2.6985          | 0.8497          |
| 2.8101 | 12   | 93.583        | 2.6846          | 0.8540          |


### Framework Versions
- Python: 3.12.11
- Sentence Transformers: 4.1.0
- Transformers: 4.50.0
- PyTorch: 2.5.1
- Accelerate: 1.7.0
- Datasets: 3.3.2
- Tokenizers: 0.21.1

## Citation

### BibTeX

#### Sentence Transformers
```bibtex
@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",
}
```

#### CoSENTLoss
```bibtex
@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}
```

<!--
## Glossary

*Clearly define terms in order to be accessible across audiences.*
-->

<!--
## Model Card Authors

*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->

<!--
## Model Card Contact

*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->