Initial release: v3-sentences

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+    "word_embedding_dimension": 1024,
+    "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
+}

README.md

@@ -0,0 +1,543 @@
+---
+tags:
+- sentence-transformers
+- sentence-similarity
+- feature-extraction
+- dense
+- generated_from_trainer
+- dataset_size:33406
+- loss:MultipleNegativesRankingLoss
+- loss:OnlineContrastiveLoss
+widget:
+- source_sentence: παρεκελεύσατο οὖν λαβοῦσαν τὸν ἐξ αὑτοῦ πεσόντα γόνον, καὶ τούτῳ
+    προσμίξασαν ἔλαιον καὶ τὸ ἀπὸ τῆς ἀκίδος ἀποστάζον αἷμα, χρῖσαι τὸν χιτῶνα τοῦ
+    Ἡρακλέους.
+  sentences:
+  - 'Då föll hans medtjänare till hans fötter och bad honom och sade: hav tålamod
+    med mig, så skall jag betala dig allt.'
+  - Vore de blott til en del eller i allmänhet bestämda, och således til en del obestämda,
+    så ägde de blott en half existents, om detta uttryck kan nyttjas, eller rättare
+    ingen.
+  - När hon försökte vinna tillbaka Herakles med en blodindränkt mantel, frätte den
+    dock bort hans hud och kött, varpå han brände sig själv på berget Oite.
+- source_sentence: ὃς γὰρ ἂν ἐπαισχυνθῇ με ἢ τοὺς ἐμοὺς λόγους ἐν τῇ γενεᾷ ταύτῃ τῇ
+    μοιχαλίδι καὶ ἁμαρτωλῷ, καὶ ὁ υἱὸς τοῦ ἀνθρώπου ἐπαισχυνθήσεται αὐτόν, ὅταν ἔλθῃ
+    ἐν τῇ δόξῃ τοῦ πατρὸς αὐτοῦ μετὰ τῶν ἀγγέλων αὐτοῦ.
+  sentences:
+  - De som förtjäna lite vill ju alltid hindra dem som förtjäna mera att kapa åt sig
+    ännu mera — och överhuvud förneka deras rätt därtill — medan de själva däremot
+    med full rätt anse sig kunna ställa anspråk.
+  - 'Alltså: Hela Kullens höjd = K.h = Pyramidens bas (P.b); och Högkullens höjd HK.h
+    = Pyramidens (P.h) höjd.'
+  - Var och en som bättrar sig på detta sätt skall jag genast gå till mötes som en
+    god fader sin felande son.
+- source_sentence: τοῦτον οὖν τὸν τίτλον πολλοὶ ἀνέγνωσαν τῶν Ἰουδαίων, ὅτι ἐγγὺς
+    ἦν ὁ τόπος τῆς πόλεως ὅπου ἐσταυρώθη ὁ Ἰησοῦς· καὶ ἦν γεγραμμένον Ἐβραϊστί, Ῥωμαϊστί,
+    Ἑλληνιστί.
+  sentences:
+  - Dessa ingredienser stamma till stor del ur Logiakällan, och det har ju förmodats,
+    att denna källa har byggt på samtida anteckningar av »publikanen Matteus», en
+    förmodad omedelbar lärjunge till Jesus.
+  - Den överskriften läste många av judarna, ty det ställe, där Jesus var korsfäst,
+    låg nära staden; och den var avfattad på hebreiska, på latin och på grekiska.
+  - Men om någon på denna grund bygger med guld, silver, dyrbara stenar, trä, hö,
+    eller strå, så skall vars och ens verk bliva uppenbart, ty dagen skall göra det
+    klart, emedan han uppenbaras med eld, och elden skall pröva, hurudant vars och
+    ens verk är.
+- source_sentence: λέγει αὐτῷ [ὁ] Πέτρος Κύριε, διὰ τί οὐ δύναμαί σοι ἀκολουθεῖν ἄρτι;
+  sentences:
+  - 'Jesus svarade honom: »Om jag icke tvår dig, så har du ingen del med mig.»'
+  - '9. Tå sade Petrus til henne: Hwarföre woren J öfwerens/ at J skullen fresta Herrans
+    Anda?'
+  - 'Petrus sade till honom: »Herre, varför kan jag icke följa dig nu?'
+- source_sentence: Οἱ οὖν στρατιῶται ὅτε ἐσταύρωσαν τὸν Ἰησοῦν ἔλαβον τὰ ἱμάτια αὐτοῦ
+    καὶ ἐποίησαν τέσσερα μέρη, ἑκάστῳ στρατιώτῃ μέρος, καὶ τὸν χιτῶνα.
+  sentences:
+  - Då nu krigsmännen hade korsfäst Jesus, togo de hans kläder och delade dem i fyra
+    delar, en del åt var krigsman.
+  - »Vi veta ju detta, att vår gamla människa har blivit korsfäst med honom, för att
+    syndakroppen skall göras om intet, så att vi icke mer tjäna synden.
+  - Derefter fatta presterna på hans vink katafalken, lyfta den på sina skuldror och
+    bära den med Simon till ett vid tvärskeppet beläget, för presterskapet afskildt
+    rum.
+pipeline_tag: sentence-similarity
+library_name: sentence-transformers
+metrics:
+- cosine_accuracy@1
+- cosine_accuracy@5
+- cosine_accuracy@10
+- cosine_accuracy@25
+- cosine_precision@1
+- cosine_precision@5
+- cosine_precision@10
+- cosine_precision@25
+- cosine_precision@100
+- cosine_recall@1
+- cosine_recall@5
+- cosine_recall@10
+- cosine_recall@25
+- cosine_recall@100
+- cosine_ndcg@10
+- cosine_mrr@10
+- cosine_map@100
+model-index:
+- name: SentenceTransformer
+  results:
+  - task:
+      type: information-retrieval
+      name: Information Retrieval
+    dataset:
+      name: labeled src to swe
+      type: labeled_src_to_swe
+    metrics:
+    - type: cosine_accuracy@1
+      value: 0.7843866171003717
+      name: Cosine Accuracy@1
+    - type: cosine_accuracy@5
+      value: 0.9368029739776952
+      name: Cosine Accuracy@5
+    - type: cosine_accuracy@10
+      value: 0.9553903345724907
+      name: Cosine Accuracy@10
+    - type: cosine_accuracy@25
+      value: 1.0
+      name: Cosine Accuracy@25
+    - type: cosine_precision@1
+      value: 0.7843866171003717
+      name: Cosine Precision@1
+    - type: cosine_precision@5
+      value: 0.18736059479553904
+      name: Cosine Precision@5
+    - type: cosine_precision@10
+      value: 0.09553903345724908
+      name: Cosine Precision@10
+    - type: cosine_precision@25
+      value: 0.04000000000000001
+      name: Cosine Precision@25
+    - type: cosine_precision@100
+      value: 0.010000000000000002
+      name: Cosine Precision@100
+    - type: cosine_recall@1
+      value: 0.7843866171003717
+      name: Cosine Recall@1
+    - type: cosine_recall@5
+      value: 0.9368029739776952
+      name: Cosine Recall@5
+    - type: cosine_recall@10
+      value: 0.9553903345724907
+      name: Cosine Recall@10
+    - type: cosine_recall@25
+      value: 1.0
+      name: Cosine Recall@25
+    - type: cosine_recall@100
+      value: 1.0
+      name: Cosine Recall@100
+    - type: cosine_ndcg@10
+      value: 0.8804480378872143
+      name: Cosine Ndcg@10
+    - type: cosine_mrr@10
+      value: 0.8553741075116541
+      name: Cosine Mrr@10
+    - type: cosine_map@100
+      value: 0.8582062450441782
+      name: Cosine Map@100
+---
+
+# SentenceTransformer
+
+This is a [sentence-transformers](https://www.SBERT.net) model trained on the triplets and pairs datasets. It maps sentences & paragraphs to a 1024-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:** [Unknown](https://huggingface.co/unknown) -->
+- **Maximum Sequence Length:** 192 tokens
+- **Output Dimensionality:** 1024 dimensions
+- **Similarity Function:** Cosine Similarity
+- **Training Datasets:**
+    - triplets
+    - pairs
+<!-- - **Language:** Unknown -->
+<!-- - **License:** Unknown -->
+
+### Model Sources
+
+- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
+- **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/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': 192, 'do_lower_case': False, 'architecture': 'XLMRobertaModel'})
+  (1): Pooling({'word_embedding_dimension': 1024, '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})
+  (2): Normalize()
+)
+```
+
+## 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 = [
+    'Οἱ οὖν στρατιῶται ὅτε ἐσταύρωσαν τὸν Ἰησοῦν ἔλαβον τὰ ἱμάτια αὐτοῦ καὶ ἐποίησαν τέσσερα μέρη, ἑκάστῳ στρατιώτῃ μέρος, καὶ τὸν χιτῶνα.',
+    'Då nu krigsmännen hade korsfäst Jesus, togo de hans kläder och delade dem i fyra delar, en del åt var krigsman.',
+    'Derefter fatta presterna på hans vink katafalken, lyfta den på sina skuldror och bära den med Simon till ett vid tvärskeppet beläget, för presterskapet afskildt rum.',
+]
+embeddings = model.encode(sentences)
+print(embeddings.shape)
+# [3, 1024]
+
+# Get the similarity scores for the embeddings
+similarities = model.similarity(embeddings, embeddings)
+print(similarities)
+# tensor([[1.0000, 0.8955, 0.3442],
+#         [0.8955, 1.0000, 0.3027],
+#         [0.3442, 0.3027, 1.0000]])
+```
+
+<!--
+### 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
+
+#### Information Retrieval
+
+* Dataset: `labeled_src_to_swe`
+* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
+
+| Metric               | Value      |
+|:---------------------|:-----------|
+| cosine_accuracy@1    | 0.7844     |
+| cosine_accuracy@5    | 0.9368     |
+| cosine_accuracy@10   | 0.9554     |
+| cosine_accuracy@25   | 1.0        |
+| cosine_precision@1   | 0.7844     |
+| cosine_precision@5   | 0.1874     |
+| cosine_precision@10  | 0.0955     |
+| cosine_precision@25  | 0.04       |
+| cosine_precision@100 | 0.01       |
+| cosine_recall@1      | 0.7844     |
+| cosine_recall@5      | 0.9368     |
+| cosine_recall@10     | 0.9554     |
+| cosine_recall@25     | 1.0        |
+| cosine_recall@100    | 1.0        |
+| **cosine_ndcg@10**   | **0.8804** |
+| cosine_mrr@10        | 0.8554     |
+| cosine_map@100       | 0.8582     |
+
+<!--
+## 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 Datasets
+
+#### triplets
+
+* Dataset: triplets
+* Size: 16,035 training samples
+* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
+* Approximate statistics based on the first 1000 samples:
+  |         | anchor                                                                              | positive                                                                           | negative                                                                           |
+  |:--------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
+  | type    | string                                                                              | string                                                                             | string                                                                             |
+  | details | <ul><li>min: 28 tokens</li><li>mean: 68.48 tokens</li><li>max: 145 tokens</li></ul> | <ul><li>min: 15 tokens</li><li>mean: 39.62 tokens</li><li>max: 93 tokens</li></ul> | <ul><li>min: 7 tokens</li><li>mean: 34.04 tokens</li><li>max: 192 tokens</li></ul> |
+* Samples:
+  | anchor                                                                                                                                             | positive                                                                                                                     | negative                                                                                                                                                                           |
+  |:---------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+  | <code>Οἱ οὖν στρατιῶται ὅτε ἐσταύρωσαν τὸν Ἰησοῦν ἔλαβον τὰ ἱμάτια αὐτοῦ καὶ ἐποίησαν τέσσερα μέρη, ἑκάστῳ στρατιώτῃ μέρος, καὶ τὸν χιτῶνα.</code> | <code>Då nu krigsmännen hade korsfäst Jesus, togo de hans kläder och delade dem i fyra delar, en del åt var krigsman.</code> | <code>Franciskanermunkarna kommo nu med sin bår och när de hade klädt honom i kåpa och rep, drogo de ned hufvan öfver hans anlete.</code>                                          |
+  | <code>Οἱ οὖν στρατιῶται ὅτε ἐσταύρωσαν τὸν Ἰησοῦν ἔλαβον τὰ ἱμάτια αὐτοῦ καὶ ἐποίησαν τέσσερα μέρη, ἑκάστῳ στρατιώτῃ μέρος, καὶ τὸν χιτῶνα.</code> | <code>Då nu krigsmännen hade korsfäst Jesus, togo de hans kläder och delade dem i fyra delar, en del åt var krigsman.</code> | <code>De togo hans krus.</code>                                                                                                                                                    |
+  | <code>Οἱ οὖν στρατιῶται ὅτε ἐσταύρωσαν τὸν Ἰησοῦν ἔλαβον τὰ ἱμάτια αὐτοῦ καὶ ἐποίησαν τέσσερα μέρη, ἑκάστῳ στρατιώτῃ μέρος, καὶ τὸν χιτῶνα.</code> | <code>Då nu krigsmännen hade korsfäst Jesus, togo de hans kläder och delade dem i fyra delar, en del åt var krigsman.</code> | <code>Derefter fatta presterna på hans vink katafalken, lyfta den på sina skuldror och bära den med Simon till ett vid tvärskeppet beläget, för presterskapet afskildt rum.</code> |
+* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
+  ```json
+  {
+      "scale": 20.0,
+      "similarity_fct": "cos_sim",
+      "gather_across_devices": false,
+      "directions": [
+          "query_to_doc"
+      ],
+      "partition_mode": "joint",
+      "hardness_mode": null,
+      "hardness_strength": 0.0
+  }
+  ```
+
+#### pairs
+
+* Dataset: pairs
+* Size: 17,371 training samples
+* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
+* Approximate statistics based on the first 1000 samples:
+  |         | sentence1                                                                          | sentence2                                                                          | label                        |
+  |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------|
+  | type    | string                                                                             | string                                                                             | int                          |
+  | details | <ul><li>min: 7 tokens</li><li>mean: 93.08 tokens</li><li>max: 192 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 45.67 tokens</li><li>max: 192 tokens</li></ul> | <ul><li>0: 100.00%</li></ul> |
+* Samples:
+  | sentence1                                                                                                        | sentence2                                                                                                                                   | label          |
+  |:-----------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------|:---------------|
+  | <code>λέγω γὰρ ὑμῖν, οὐ μὴ πίω ἀπὸ τοῦ νῦν ἀπὸ τοῦ γενήματος τῆς ἀμπέλου ἕως οὗ ἡ βασιλεία τοῦ θεοῦ ἔλθῃ.</code> | <code>ty jag säger eder att jag icke mer skall fira denna högtid, förrän den kommer till fullbordan i Guds rike.»</code>                    | <code>0</code> |
+  | <code>ἐξῆλθεν οὖν ὁ μαθητὴς ὁ ἄλλος ὁ γνωστὸς τοῦ ἀρχιερέως καὶ εἶπεν τῇ θυρωρῷ καὶ εἰσήγαγεν τὸν Πέτρον.</code> | <code>Den lärjungen var bekant med översteprästen och gick med Jesus in på översteprästens gård; men Petrus stod utanför vid porten.</code> | <code>0</code> |
+  | <code>ὥστε οἱ ἐκ πίστεως εὐλογοῦνται σὺν τῷ πιστῷ Ἀβραάμ.</code>                                                 | <code>Veten därför, att de som äro av tron, de äro Abrahams barn.</code>                                                                    | <code>0</code> |
+* Loss: [<code>OnlineContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss)
+
+### Training Hyperparameters
+#### Non-Default Hyperparameters
+
+- `per_device_train_batch_size`: 32
+- `num_train_epochs`: 1.0
+- `learning_rate`: 2e-06
+- `warmup_steps`: 0.05
+- `bf16`: True
+- `eval_strategy`: steps
+- `per_device_eval_batch_size`: 128
+- `load_best_model_at_end`: True
+- `dataloader_drop_last`: True
+- `dataloader_num_workers`: 4
+- `ddp_find_unused_parameters`: True
+- `batch_sampler`: no_duplicates
+- `multi_dataset_batch_sampler`: round_robin
+
+#### All Hyperparameters
+<details><summary>Click to expand</summary>
+
+- `per_device_train_batch_size`: 32
+- `num_train_epochs`: 1.0
+- `max_steps`: -1
+- `learning_rate`: 2e-06
+- `lr_scheduler_type`: linear
+- `lr_scheduler_kwargs`: None
+- `warmup_steps`: 0.05
+- `optim`: adamw_torch_fused
+- `optim_args`: None
+- `weight_decay`: 0.0
+- `adam_beta1`: 0.9
+- `adam_beta2`: 0.999
+- `adam_epsilon`: 1e-08
+- `optim_target_modules`: None
+- `gradient_accumulation_steps`: 1
+- `average_tokens_across_devices`: True
+- `max_grad_norm`: 1.0
+- `label_smoothing_factor`: 0.0
+- `bf16`: True
+- `fp16`: False
+- `bf16_full_eval`: False
+- `fp16_full_eval`: False
+- `tf32`: None
+- `gradient_checkpointing`: False
+- `gradient_checkpointing_kwargs`: None
+- `torch_compile`: False
+- `torch_compile_backend`: None
+- `torch_compile_mode`: None
+- `use_liger_kernel`: False
+- `liger_kernel_config`: None
+- `use_cache`: False
+- `neftune_noise_alpha`: None
+- `torch_empty_cache_steps`: None
+- `auto_find_batch_size`: False
+- `log_on_each_node`: True
+- `logging_nan_inf_filter`: True
+- `include_num_input_tokens_seen`: no
+- `log_level`: passive
+- `log_level_replica`: warning
+- `disable_tqdm`: False
+- `project`: huggingface
+- `trackio_space_id`: trackio
+- `eval_strategy`: steps
+- `per_device_eval_batch_size`: 128
+- `prediction_loss_only`: True
+- `eval_on_start`: False
+- `eval_do_concat_batches`: True
+- `eval_use_gather_object`: False
+- `eval_accumulation_steps`: None
+- `include_for_metrics`: []
+- `batch_eval_metrics`: False
+- `save_only_model`: False
+- `save_on_each_node`: False
+- `enable_jit_checkpoint`: False
+- `push_to_hub`: False
+- `hub_private_repo`: None
+- `hub_model_id`: None
+- `hub_strategy`: every_save
+- `hub_always_push`: False
+- `hub_revision`: None
+- `load_best_model_at_end`: True
+- `ignore_data_skip`: False
+- `restore_callback_states_from_checkpoint`: False
+- `full_determinism`: False
+- `seed`: 42
+- `data_seed`: None
+- `use_cpu`: False
+- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
+- `parallelism_config`: None
+- `dataloader_drop_last`: True
+- `dataloader_num_workers`: 4
+- `dataloader_pin_memory`: True
+- `dataloader_persistent_workers`: False
+- `dataloader_prefetch_factor`: None
+- `remove_unused_columns`: True
+- `label_names`: None
+- `train_sampling_strategy`: random
+- `length_column_name`: length
+- `ddp_find_unused_parameters`: True
+- `ddp_bucket_cap_mb`: None
+- `ddp_broadcast_buffers`: False
+- `ddp_backend`: None
+- `ddp_timeout`: 1800
+- `fsdp`: []
+- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
+- `deepspeed`: None
+- `debug`: []
+- `skip_memory_metrics`: True
+- `do_predict`: False
+- `resume_from_checkpoint`: None
+- `warmup_ratio`: None
+- `local_rank`: -1
+- `prompts`: None
+- `batch_sampler`: no_duplicates
+- `multi_dataset_batch_sampler`: round_robin
+- `router_mapping`: {}
+- `learning_rate_mapping`: {}
+
+</details>
+
+### Training Logs
+| Epoch   | Step    | Training Loss | labeled_src_to_swe_cosine_ndcg@10 |
+|:-------:|:-------:|:-------------:|:---------------------------------:|
+| -1      | -1      | -             | 0.8786                            |
+| 0.04    | 10      | 0.0745        | -                                 |
+| 0.08    | 20      | 0.0752        | -                                 |
+| 0.1     | 25      | -             | 0.8772                            |
+| 0.12    | 30      | 0.0674        | -                                 |
+| 0.16    | 40      | 0.0577        | -                                 |
+| 0.2     | 50      | 0.0564        | 0.8777                            |
+| 0.24    | 60      | 0.0498        | -                                 |
+| 0.28    | 70      | 0.0479        | -                                 |
+| 0.3     | 75      | -             | 0.8796                            |
+| 0.32    | 80      | 0.0439        | -                                 |
+| 0.36    | 90      | 0.0432        | -                                 |
+| 0.4     | 100     | 0.0406        | 0.8796                            |
+| 0.44    | 110     | 0.0400        | -                                 |
+| 0.48    | 120     | 0.0356        | -                                 |
+| 0.5     | 125     | -             | 0.8796                            |
+| 0.52    | 130     | 0.0417        | -                                 |
+| 0.56    | 140     | 0.0421        | -                                 |
+| 0.6     | 150     | 0.0534        | 0.8796                            |
+| 0.64    | 160     | 0.0335        | -                                 |
+| 0.68    | 170     | 0.0370        | -                                 |
+| **0.7** | **175** | **-**         | **0.8809**                        |
+| 0.72    | 180     | 0.0385        | -                                 |
+| 0.76    | 190     | 0.0326        | -                                 |
+| 0.8     | 200     | 0.0344        | 0.8804                            |
+| 0.84    | 210     | 0.0337        | -                                 |
+| 0.88    | 220     | 0.0313        | -                                 |
+| 0.9     | 225     | -             | 0.8804                            |
+| 0.92    | 230     | 0.0335        | -                                 |
+| 0.96    | 240     | 0.0328        | -                                 |
+
+* The bold row denotes the saved checkpoint.
+
+### Framework Versions
+- Python: 3.12.3
+- Sentence Transformers: 5.3.0
+- Transformers: 5.3.0
+- PyTorch: 2.9.0+cu128
+- Accelerate: 1.13.0
+- Datasets: 4.8.5
+- Tokenizers: 0.22.2
+
+## 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",
+}
+```
+
+#### MultipleNegativesRankingLoss
+```bibtex
+@misc{oord2019representationlearningcontrastivepredictive,
+      title={Representation Learning with Contrastive Predictive Coding},
+      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
+      year={2019},
+      eprint={1807.03748},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG},
+      url={https://arxiv.org/abs/1807.03748},
+}
+```
+
+<!--
+## 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.*
+-->

config.json

@@ -0,0 +1,30 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "XLMRobertaModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "is_decoder": false,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 8194,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.3.0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "__version__": {
+    "sentence_transformers": "5.3.0",
+    "transformers": "5.3.0",
+    "pytorch": "2.9.0+cu128"
+  },
+  "model_type": "SentenceTransformer",
+  "prompts": {
+    "query": "",
+    "document": ""
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afaeaeab7a564b3714920717ae627a4d01639df919707e86d2ca6be3eee0f2b7
+size 2271064408

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+    "max_seq_length": 192,
+    "do_lower_case": false
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:826dde65dfa2758657bbb0543df764e99dd9bb616f972dac0240af0a226b0746
+size 17098338

tokenizer_config.json

@@ -0,0 +1,23 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": true,
+  "mask_token": "<mask>",
+  "max_length": 256,
+  "model_max_length": 192,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "stride": 0,
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<unk>"
+}