--- language: - ar library_name: sentence-transformers tags: - mteb - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:557850 - loss:MatryoshkaLoss - loss:MultipleNegativesRankingLoss base_model: UBC-NLP/MARBERTv2 datasets: - Omartificial-Intelligence-Space/Arabic-NLi-Triplet metrics: - pearson_cosine - spearman_cosine - pearson_manhattan - spearman_manhattan - pearson_euclidean - spearman_euclidean - pearson_dot - spearman_dot - pearson_max - spearman_max widget: - source_sentence: ذكر متوازن بعناية يقف على قدم واحدة بالقرب من منطقة شاطئ المحيط النظيفة sentences: - رجل يقدم عرضاً - هناك رجل بالخارج قرب الشاطئ - رجل يجلس على أريكه - source_sentence: رجل يقفز إلى سريره القذر sentences: - السرير قذر. - رجل يضحك أثناء غسيل الملابس - الرجل على القمر - source_sentence: الفتيات بالخارج sentences: - امرأة تلف الخيط إلى كرات بجانب كومة من الكرات - فتيان يركبان في جولة متعة - >- ثلاث فتيات يقفون سوية في غرفة واحدة تستمع وواحدة تكتب على الحائط والثالثة تتحدث إليهن - source_sentence: الرجل يرتدي قميصاً أزرق. sentences: - >- رجل يرتدي قميصاً أزرق يميل إلى الجدار بجانب الطريق مع شاحنة زرقاء وسيارة حمراء مع الماء في الخلفية. - كتاب القصص مفتوح - رجل يرتدي قميص أسود يعزف على الجيتار. - source_sentence: يجلس شاب ذو شعر أشقر على الحائط يقرأ جريدة بينما تمر امرأة وفتاة شابة. sentences: - ذكر شاب ينظر إلى جريدة بينما تمر إمرأتان بجانبه - رجل يستلقي على وجهه على مقعد في الحديقة. - الشاب نائم بينما الأم تقود ابنتها إلى الحديقة pipeline_tag: sentence-similarity model-index: - name: Omartificial-Intelligence-Space/Marbert-all-nli-triplet-Matryoshka results: - dataset: config: default name: MTEB BIOSSES (default) revision: d3fb88f8f02e40887cd149695127462bbcf29b4a split: test type: mteb/biosses-sts metrics: - type: cosine_pearson value: 49.25240527202211 - type: cosine_spearman value: 51.87708566904703 - type: euclidean_pearson value: 49.790877425774696 - type: euclidean_spearman value: 51.725274981021855 - type: main_score value: 51.87708566904703 - type: manhattan_pearson value: 52.31560776967401 - type: manhattan_spearman value: 54.28979124658997 task: type: STS - dataset: config: default name: MTEB SICK-R (default) revision: 20a6d6f312dd54037fe07a32d58e5e168867909d split: test type: mteb/sickr-sts metrics: - type: cosine_pearson value: 65.81089479351829 - type: cosine_spearman value: 65.80163441928238 - type: euclidean_pearson value: 65.2718874370746 - type: euclidean_spearman value: 65.92429031695988 - type: main_score value: 65.80163441928238 - type: manhattan_pearson value: 65.28701419332383 - type: manhattan_spearman value: 65.94229793651319 task: type: STS - dataset: config: default name: MTEB STS12 (default) revision: a0d554a64d88156834ff5ae9920b964011b16384 split: test type: mteb/sts12-sts metrics: - type: cosine_pearson value: 65.11346939995998 - type: cosine_spearman value: 63.00297824477175 - type: euclidean_pearson value: 63.85320097970942 - type: euclidean_spearman value: 63.25151047701848 - type: main_score value: 63.00297824477175 - type: manhattan_pearson value: 64.40291990853984 - type: manhattan_spearman value: 63.63497232399945 task: type: STS - dataset: config: default name: MTEB STS13 (default) revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca split: test type: mteb/sts13-sts metrics: - type: cosine_pearson value: 52.2735823521702 - type: cosine_spearman value: 52.23198766098021 - type: euclidean_pearson value: 54.12467577456837 - type: euclidean_spearman value: 52.40014028261351 - type: main_score value: 52.23198766098021 - type: manhattan_pearson value: 54.38052509834607 - type: manhattan_spearman value: 52.70836595958237 task: type: STS - dataset: config: default name: MTEB STS14 (default) revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375 split: test type: mteb/sts14-sts metrics: - type: cosine_pearson value: 58.55307076840419 - type: cosine_spearman value: 59.2261024017655 - type: euclidean_pearson value: 59.55734715751804 - type: euclidean_spearman value: 60.135899681574834 - type: main_score value: 59.2261024017655 - type: manhattan_pearson value: 59.99274396356966 - type: manhattan_spearman value: 60.44325356503041 task: type: STS - dataset: config: default name: MTEB STS15 (default) revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3 split: test type: mteb/sts15-sts metrics: - type: cosine_pearson value: 68.94418532602707 - type: cosine_spearman value: 70.01912156519296 - type: euclidean_pearson value: 71.67028435860581 - type: euclidean_spearman value: 71.48252471922122 - type: main_score value: 70.01912156519296 - type: manhattan_pearson value: 71.9587452337792 - type: manhattan_spearman value: 71.69160519065173 task: type: STS - dataset: config: default name: MTEB STS16 (default) revision: 4d8694f8f0e0100860b497b999b3dbed754a0513 split: test type: mteb/sts16-sts metrics: - type: cosine_pearson value: 62.81619254162203 - type: cosine_spearman value: 64.98814526698425 - type: euclidean_pearson value: 66.43531796610995 - type: euclidean_spearman value: 66.53768451143964 - type: main_score value: 64.98814526698425 - type: manhattan_pearson value: 66.57822125651369 - type: manhattan_spearman value: 66.71830390508079 task: type: STS - dataset: config: ar-ar name: MTEB STS17 (ar-ar) revision: faeb762787bd10488a50c8b5be4a3b82e411949c split: test type: mteb/sts17-crosslingual-sts metrics: - type: cosine_pearson value: 81.68055610903552 - type: cosine_spearman value: 82.18125783448961 - type: euclidean_pearson value: 80.5422740473486 - type: euclidean_spearman value: 81.79456727036232 - type: main_score value: 82.18125783448961 - type: manhattan_pearson value: 80.43564733654793 - type: manhattan_spearman value: 81.76103816207625 task: type: STS - dataset: config: ar name: MTEB STS22 (ar) revision: de9d86b3b84231dc21f76c7b7af1f28e2f57f6e3 split: test type: mteb/sts22-crosslingual-sts metrics: - type: cosine_pearson value: 51.33460593849487 - type: cosine_spearman value: 58.07741072443786 - type: euclidean_pearson value: 54.26430308336828 - type: euclidean_spearman value: 58.8384539429318 - type: main_score value: 58.07741072443786 - type: manhattan_pearson value: 54.41587176266624 - type: manhattan_spearman value: 58.831993325957086 task: type: STS - dataset: config: default name: MTEB STSBenchmark (default) revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831 split: test type: mteb/stsbenchmark-sts metrics: - type: cosine_pearson value: 61.11956207522431 - type: cosine_spearman value: 61.16768766134144 - type: euclidean_pearson value: 64.44141934993837 - type: euclidean_spearman value: 63.450379593077066 - type: main_score value: 61.16768766134144 - type: manhattan_pearson value: 64.43852352892529 - type: manhattan_spearman value: 63.57630045107761 task: type: STS - dataset: config: default name: MTEB SummEval (default) revision: cda12ad7615edc362dbf25a00fdd61d3b1eaf93c split: test type: mteb/summeval metrics: - type: cosine_pearson value: 29.583566160417668 - type: cosine_spearman value: 29.534419950502212 - type: dot_pearson value: 28.13970643170574 - type: dot_spearman value: 28.907762267009073 - type: main_score value: 29.534419950502212 - type: pearson value: 29.583566160417668 - type: spearman value: 29.534419950502212 task: type: Summarization - name: SentenceTransformer based on UBC-NLP/MARBERTv2 results: - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts test 768 type: sts-test-768 metrics: - type: pearson_cosine value: 0.611168498883907 name: Pearson Cosine - type: spearman_cosine value: 0.6116733587939157 name: Spearman Cosine - type: pearson_manhattan value: 0.6443687886661206 name: Pearson Manhattan - type: spearman_manhattan value: 0.6358107360369792 name: Spearman Manhattan - type: pearson_euclidean value: 0.644404066642609 name: Pearson Euclidean - type: spearman_euclidean value: 0.6345893921062774 name: Spearman Euclidean - type: pearson_dot value: 0.4723643245352202 name: Pearson Dot - type: spearman_dot value: 0.44844519905410135 name: Spearman Dot - type: pearson_max value: 0.644404066642609 name: Pearson Max - type: spearman_max value: 0.6358107360369792 name: Spearman Max - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts test 512 type: sts-test-512 metrics: - type: pearson_cosine value: 0.6664570291720014 name: Pearson Cosine - type: spearman_cosine value: 0.6647687532159875 name: Spearman Cosine - type: pearson_manhattan value: 0.6429976947418544 name: Pearson Manhattan - type: spearman_manhattan value: 0.6334753432753939 name: Spearman Manhattan - type: pearson_euclidean value: 0.6466249455585532 name: Pearson Euclidean - type: spearman_euclidean value: 0.6373181315122213 name: Spearman Euclidean - type: pearson_dot value: 0.5370129457359227 name: Pearson Dot - type: spearman_dot value: 0.5241649973373772 name: Spearman Dot - type: pearson_max value: 0.6664570291720014 name: Pearson Max - type: spearman_max value: 0.6647687532159875 name: Spearman Max - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts test 256 type: sts-test-256 metrics: - type: pearson_cosine value: 0.6601248277308522 name: Pearson Cosine - type: spearman_cosine value: 0.6592739654246011 name: Spearman Cosine - type: pearson_manhattan value: 0.6361644543165994 name: Pearson Manhattan - type: spearman_manhattan value: 0.6250621947417249 name: Spearman Manhattan - type: pearson_euclidean value: 0.6408426652431157 name: Pearson Euclidean - type: spearman_euclidean value: 0.6300109524350457 name: Spearman Euclidean - type: pearson_dot value: 0.5250513197384045 name: Pearson Dot - type: spearman_dot value: 0.5154779060125071 name: Spearman Dot - type: pearson_max value: 0.6601248277308522 name: Pearson Max - type: spearman_max value: 0.6592739654246011 name: Spearman Max - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts test 128 type: sts-test-128 metrics: - type: pearson_cosine value: 0.6549481034721005 name: Pearson Cosine - type: spearman_cosine value: 0.6523201621940143 name: Spearman Cosine - type: pearson_manhattan value: 0.6342700090917214 name: Pearson Manhattan - type: spearman_manhattan value: 0.6226791710099966 name: Spearman Manhattan - type: pearson_euclidean value: 0.6397224689512541 name: Pearson Euclidean - type: spearman_euclidean value: 0.6280973341704362 name: Spearman Euclidean - type: pearson_dot value: 0.47240889358810917 name: Pearson Dot - type: spearman_dot value: 0.4633669926372942 name: Spearman Dot - type: pearson_max value: 0.6549481034721005 name: Pearson Max - type: spearman_max value: 0.6523201621940143 name: Spearman Max - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts test 64 type: sts-test-64 metrics: - type: pearson_cosine value: 0.6367217585211098 name: Pearson Cosine - type: spearman_cosine value: 0.6370191671711296 name: Spearman Cosine - type: pearson_manhattan value: 0.6263730801254332 name: Pearson Manhattan - type: spearman_manhattan value: 0.6118927366012856 name: Spearman Manhattan - type: pearson_euclidean value: 0.6327699647617465 name: Pearson Euclidean - type: spearman_euclidean value: 0.6180184829867724 name: Spearman Euclidean - type: pearson_dot value: 0.41169381399943167 name: Pearson Dot - type: spearman_dot value: 0.40444222536491986 name: Spearman Dot - type: pearson_max value: 0.6367217585211098 name: Pearson Max - type: spearman_max value: 0.6370191671711296 name: Spearman Max license: apache-2.0 --- # SentenceTransformer based on UBC-NLP/MARBERTv2 This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [UBC-NLP/MARBERTv2](https://huggingface.co/UBC-NLP/MARBERTv2) on the Omartificial-Intelligence-Space/arabic-n_li-triplet 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:** [UBC-NLP/MARBERTv2](https://huggingface.co/UBC-NLP/MARBERTv2) - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 768 tokens - **Similarity Function:** Cosine Similarity - **Training Dataset:** - Omartificial-Intelligence-Space/arabic-n_li-triplet ### 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("Omartificial-Intelligence-Space/Marbert-all-nli-triplet") # Run inference sentences = [ 'يجلس شاب ذو شعر أشقر على الحائط يقرأ جريدة بينما تمر امرأة وفتاة شابة.', 'ذكر شاب ينظر إلى جريدة بينما تمر إمرأتان بجانبه', 'الشاب نائم بينما الأم تقود ابنتها إلى الحديقة', ] 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] ``` ## Evaluation ### Metrics #### Semantic Similarity * Dataset: `sts-test-768` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | Value | |:--------------------|:-----------| | pearson_cosine | 0.6112 | | **spearman_cosine** | **0.6117** | | pearson_manhattan | 0.6444 | | spearman_manhattan | 0.6358 | | pearson_euclidean | 0.6444 | | spearman_euclidean | 0.6346 | | pearson_dot | 0.4724 | | spearman_dot | 0.4484 | | pearson_max | 0.6444 | | spearman_max | 0.6358 | #### Semantic Similarity * Dataset: `sts-test-512` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | Value | |:--------------------|:-----------| | pearson_cosine | 0.6665 | | **spearman_cosine** | **0.6648** | | pearson_manhattan | 0.643 | | spearman_manhattan | 0.6335 | | pearson_euclidean | 0.6466 | | spearman_euclidean | 0.6373 | | pearson_dot | 0.537 | | spearman_dot | 0.5242 | | pearson_max | 0.6665 | | spearman_max | 0.6648 | #### Semantic Similarity * Dataset: `sts-test-256` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | Value | |:--------------------|:-----------| | pearson_cosine | 0.6601 | | **spearman_cosine** | **0.6593** | | pearson_manhattan | 0.6362 | | spearman_manhattan | 0.6251 | | pearson_euclidean | 0.6408 | | spearman_euclidean | 0.63 | | pearson_dot | 0.5251 | | spearman_dot | 0.5155 | | pearson_max | 0.6601 | | spearman_max | 0.6593 | #### Semantic Similarity * Dataset: `sts-test-128` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | Value | |:--------------------|:-----------| | pearson_cosine | 0.6549 | | **spearman_cosine** | **0.6523** | | pearson_manhattan | 0.6343 | | spearman_manhattan | 0.6227 | | pearson_euclidean | 0.6397 | | spearman_euclidean | 0.6281 | | pearson_dot | 0.4724 | | spearman_dot | 0.4634 | | pearson_max | 0.6549 | | spearman_max | 0.6523 | #### Semantic Similarity * Dataset: `sts-test-64` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | Value | |:--------------------|:----------| | pearson_cosine | 0.6367 | | **spearman_cosine** | **0.637** | | pearson_manhattan | 0.6264 | | spearman_manhattan | 0.6119 | | pearson_euclidean | 0.6328 | | spearman_euclidean | 0.618 | | pearson_dot | 0.4117 | | spearman_dot | 0.4044 | | pearson_max | 0.6367 | | spearman_max | 0.637 | ## Training Details ### Training Dataset #### Omartificial-Intelligence-Space/arabic-n_li-triplet * Dataset: Omartificial-Intelligence-Space/arabic-n_li-triplet * Size: 557,850 training samples * Columns: anchor, positive, and negative * Approximate statistics based on the first 1000 samples: | | anchor | positive | negative | |:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | anchor | positive | negative | |:------------------------------------------------------------|:--------------------------------------------|:------------------------------------| | شخص على حصان يقفز فوق طائرة معطلة | شخص في الهواء الطلق، على حصان. | شخص في مطعم، يطلب عجة. | | أطفال يبتسمون و يلوحون للكاميرا | هناك أطفال حاضرون | الاطفال يتجهمون | | صبي يقفز على لوح التزلج في منتصف الجسر الأحمر. | الفتى يقوم بخدعة التزلج | الصبي يتزلج على الرصيف | * Loss: [MatryoshkaLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters: ```json { "loss": "MultipleNegativesRankingLoss", "matryoshka_dims": [ 768, 512, 256, 128, 64 ], "matryoshka_weights": [ 1, 1, 1, 1, 1 ], "n_dims_per_step": -1 } ``` ### Evaluation Dataset #### Omartificial-Intelligence-Space/arabic-n_li-triplet * Dataset: Omartificial-Intelligence-Space/arabic-n_li-triplet * Size: 6,584 evaluation samples * Columns: anchor, positive, and negative * Approximate statistics based on the first 1000 samples: | | anchor | positive | negative | |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | anchor | positive | negative | |:-----------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------|:---------------------------------------------------| | امرأتان يتعانقان بينما يحملان حزمة | إمرأتان يحملان حزمة | الرجال يتشاجرون خارج مطعم | | طفلين صغيرين يرتديان قميصاً أزرق، أحدهما يرتدي الرقم 9 والآخر يرتدي الرقم 2 يقفان على خطوات خشبية في الحمام ويغسلان أيديهما في المغسلة. | طفلين يرتديان قميصاً مرقماً يغسلون أيديهم | طفلين يرتديان سترة يذهبان إلى المدرسة | | رجل يبيع الدونات لعميل خلال معرض عالمي أقيم في مدينة أنجليس | رجل يبيع الدونات لعميل | امرأة تشرب قهوتها في مقهى صغير | * Loss: [MatryoshkaLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters: ```json { "loss": "MultipleNegativesRankingLoss", "matryoshka_dims": [ 768, 512, 256, 128, 64 ], "matryoshka_weights": [ 1, 1, 1, 1, 1 ], "n_dims_per_step": -1 } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `num_train_epochs`: 1 - `warmup_ratio`: 0.1 - `fp16`: True - `batch_sampler`: no_duplicates #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-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`: 1 - `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 - `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`: True - `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, '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`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `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_sampler`: no_duplicates - `multi_dataset_batch_sampler`: proportional
### Training Logs | Epoch | Step | Training Loss | sts-test-128_spearman_cosine | sts-test-256_spearman_cosine | sts-test-512_spearman_cosine | sts-test-64_spearman_cosine | sts-test-768_spearman_cosine | |:------:|:----:|:-------------:|:----------------------------:|:----------------------------:|:----------------------------:|:---------------------------:|:----------------------------:| | 0.0229 | 200 | 25.0771 | - | - | - | - | - | | 0.0459 | 400 | 9.1435 | - | - | - | - | - | | 0.0688 | 600 | 8.0492 | - | - | - | - | - | | 0.0918 | 800 | 7.1378 | - | - | - | - | - | | 0.1147 | 1000 | 7.6249 | - | - | - | - | - | | 0.1377 | 1200 | 7.3604 | - | - | - | - | - | | 0.1606 | 1400 | 6.5783 | - | - | - | - | - | | 0.1835 | 1600 | 6.4145 | - | - | - | - | - | | 0.2065 | 1800 | 6.1781 | - | - | - | - | - | | 0.2294 | 2000 | 6.2375 | - | - | - | - | - | | 0.2524 | 2200 | 6.2587 | - | - | - | - | - | | 0.2753 | 2400 | 6.0826 | - | - | - | - | - | | 0.2983 | 2600 | 6.1514 | - | - | - | - | - | | 0.3212 | 2800 | 5.6949 | - | - | - | - | - | | 0.3442 | 3000 | 6.0062 | - | - | - | - | - | | 0.3671 | 3200 | 5.7551 | - | - | - | - | - | | 0.3900 | 3400 | 5.658 | - | - | - | - | - | | 0.4130 | 3600 | 5.7135 | - | - | - | - | - | | 0.4359 | 3800 | 5.3909 | - | - | - | - | - | | 0.4589 | 4000 | 5.5068 | - | - | - | - | - | | 0.4818 | 4200 | 5.2261 | - | - | - | - | - | | 0.5048 | 4400 | 5.1674 | - | - | - | - | - | | 0.5277 | 4600 | 5.0427 | - | - | - | - | - | | 0.5506 | 4800 | 5.3824 | - | - | - | - | - | | 0.5736 | 5000 | 5.3063 | - | - | - | - | - | | 0.5965 | 5200 | 5.2174 | - | - | - | - | - | | 0.6195 | 5400 | 5.2116 | - | - | - | - | - | | 0.6424 | 5600 | 5.2226 | - | - | - | - | - | | 0.6654 | 5800 | 5.2051 | - | - | - | - | - | | 0.6883 | 6000 | 5.204 | - | - | - | - | - | | 0.7113 | 6200 | 5.154 | - | - | - | - | - | | 0.7342 | 6400 | 5.0236 | - | - | - | - | - | | 0.7571 | 6600 | 4.9476 | - | - | - | - | - | | 0.7801 | 6800 | 4.0164 | - | - | - | - | - | | 0.8030 | 7000 | 3.5707 | - | - | - | - | - | | 0.8260 | 7200 | 3.3586 | - | - | - | - | - | | 0.8489 | 7400 | 3.2376 | - | - | - | - | - | | 0.8719 | 7600 | 3.0282 | - | - | - | - | - | | 0.8948 | 7800 | 2.901 | - | - | - | - | - | | 0.9177 | 8000 | 2.9371 | - | - | - | - | - | | 0.9407 | 8200 | 2.8362 | - | - | - | - | - | | 0.9636 | 8400 | 2.8121 | - | - | - | - | - | | 0.9866 | 8600 | 2.7105 | - | - | - | - | - | | 1.0 | 8717 | - | 0.6523 | 0.6593 | 0.6648 | 0.6370 | 0.6117 | ### Framework Versions - Python: 3.9.18 - Sentence Transformers: 3.0.1 - Transformers: 4.40.0 - PyTorch: 2.2.2+cu121 - Accelerate: 0.26.1 - Datasets: 2.19.0 - Tokenizers: 0.19.1 ## 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", } ``` #### MatryoshkaLoss ```bibtex @misc{kusupati2024matryoshka, title={Matryoshka Representation Learning}, author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi}, year={2024}, eprint={2205.13147}, archivePrefix={arXiv}, primaryClass={cs.LG} } ``` #### MultipleNegativesRankingLoss ```bibtex @misc{henderson2017efficient, title={Efficient Natural Language Response Suggestion for Smart Reply}, author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil}, year={2017}, eprint={1705.00652}, archivePrefix={arXiv}, primaryClass={cs.CL} } ``` ## Acknowledgments The author would like to thank Prince Sultan University for their invaluable support in this project. Their contributions and resources have been instrumental in the development and fine-tuning of these models. ```markdown ## Citation If you use the Arabic Matryoshka Embeddings Model, please cite it as follows: ```bibtex @misc{nacar2024enhancingsemanticsimilarityunderstanding, title={Enhancing Semantic Similarity Understanding in Arabic NLP with Nested Embedding Learning}, author={Omer Nacar and Anis Koubaa}, year={2024}, eprint={2407.21139}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2407.21139}, }