nphach/jp-parallel-gloss
0119
1---2tags:3- sentence-transformers4- sentence-similarity5- feature-extraction6- generated_from_trainer7- dataset_size:44048448- loss:CoSENTLoss9base_model: sentence-transformers/all-MiniLM-L6-v210pipeline_tag: sentence-similarity11library_name: sentence-transformers12language: en13metrics:14- cosine_accuracy15- cosine_accuracy_threshold16- cosine_f117- cosine_f1_threshold18- cosine_precision19- cosine_recall20- cosine_ap21- cosine_mcc22model-index:23- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v224 results:25 - task:26 type: binary-classification27 name: Binary Classification28 metrics:29 - type: cosine_accuracy30 value: 0.989754595080266431 name: Cosine Accuracy32 - type: cosine_accuracy_threshold33 value: 0.433196246623992934 name: Cosine Accuracy Threshold35 - type: cosine_f136 value: 0.968556578320901537 name: Cosine F138 - type: cosine_f1_threshold39 value: 0.432469695806503340 name: Cosine F1 Threshold41 - type: cosine_precision42 value: 0.969672293942403243 name: Cosine Precision44 - type: cosine_recall45 value: 0.967443427257955846 name: Cosine Recall47 - type: cosine_ap48 value: 0.993400870135188449 name: Cosine Ap50 - type: cosine_mcc51 value: 0.962437782460890152 name: Cosine Mcc53---54 55# jp-parallel-gloss56 57jp-parallel-gloss makes predictions on similarity of Japanese-to-English glosses (definitions).58This is a [sentence-transformers](https://www.SBERT.net) model fine-tuned using a dataset of 4M+ parallel/non-parallel gloss pairs from the [JMDict](https://www.edrdg.org/wiki/index.php/JMdict-EDICT_Dictionary_Project) database and antonym/synonym pairs from [WordNet](https://wordnet.princeton.edu/). The base model used is `cross-encoder/ms-macro-MiniLM-L-6-v2`. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.59See its application in [Kotoba Tag](https://github.com/nphach/kotoba-tag/)60 61## Model Details62 63### Model Description64- **Model Type:** Sentence Transformer65- **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision fa97f6e7cb1a59073dff9e6b13e2715cf7475ac9 -->66- **Maximum Sequence Length:** 256 tokens67- **Output Dimensionality:** 384 dimensions68- **Similarity Function:** Cosine Similarity69- **Language:** English70 71### Model Sources72 73- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)74- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)75- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)76 77### Full Model Architecture78 79```80SentenceTransformer(81 (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel 82 (1): Pooling({'word_embedding_dimension': 384, '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})83 (2): Normalize()84)85```86 87## Usage88 89First install the Sentence Transformers library:90 91```bash92pip install -U sentence-transformers93```94 95Then you can load this model and run inference.96```python97from sentence_transformers import SentenceTransformer98 99# Download from the 🤗 Hub100model = SentenceTransformer("sentence_transformers_model_id")101# Run inference102sentences = [103 'dearest',104 'to become verminous',105 "having an (overly) strong attachment to one's mother",106]107embeddings = model.encode(sentences)108print(embeddings.shape)109# [3, 384]110 111# Get the similarity scores for the embeddings112similarities = model.similarity(embeddings, embeddings)113print(similarities.shape)114# [3, 3]115```116 117## Evaluation118 119### Metrics120 121#### Binary Classification122* Evaluated with [<code>BinaryClassificationEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator)123 124| Metric | Value |125|:--------------------------|:-------------------|126| cosine_accuracy | 0.9897545950802664 |127| cosine_accuracy_threshold | 0.4331962466239929 |128| cosine_f1 | 0.9685565783209015 |129| cosine_f1_threshold | 0.4324696958065033 |130| cosine_precision | 0.9696722939424032 |131| cosine_recall | 0.9674434272579558 |132| cosine_ap | 0.9934008701351884 |133| cosine_mcc | 0.9624377824608901 |134 135## Training Details136 137* Size: 4,404,844 training samples138* Columns: <code>text1</code>, <code>text2</code>, and <code>label</code>139* Approximate statistics based on the first 1000 samples:140 | | text1 | text2 | label |141 |:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:------------------------------------------------------|142 | type | string | string | int |143 | details | <ul><li>min: 3 tokens</li><li>mean: 5.65 tokens</li><li>max: 27 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 5.64 tokens</li><li>max: 31 tokens</li></ul> | <ul><li>False: ~91.80%</li><li>True: ~8.20%</li></ul> |144* Samples:145 | text1 | text2 | label |146 |:-------------------------------|:--------------------------------------|:-------------------|147 | <code>based on</code> | <code>making up (a deficiency)</code> | <code>False</code> |148 | <code>folk (esp. music)</code> | <code>if possible</code> | <code>False</code> |149 | <code>to start</code> | <code>to die</code> | <code>False</code> |150* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:151 ```json152 {153 "scale": 20.0,154 "similarity_fct": "pairwise_cos_sim"155 }156 ```157 158### Evaluation159 160* Size: 550,605 evaluation samples161* Columns: <code>text1</code>, <code>text2</code>, and <code>label</code>162* Approximate statistics based on the first 1000 samples:163 | | text1 | text2 | label |164 |:--------|:---------------------------------------------------------------------------------|:--------------------------------------------------------------------------------|:------------------------------------------------------|165 | type | string | string | int |166 | details | <ul><li>min: 3 tokens</li><li>mean: 5.74 tokens</li><li>max: 28 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 5.7 tokens</li><li>max: 32 tokens</li></ul> | <ul><li>False: ~91.60%</li><li>True: ~8.40%</li></ul> |167* Samples:168 | text1 | text2 | label |169 |:----------------------------------------------------------------------------------|:-------------------------------------|:-------------------|170 | <code>taking one's children along (to an event, into a new marriage, etc.)</code> | <code>disconnect</code> | <code>False</code> |171 | <code>to thunder</code> | <code>sheet</code> | <code>False</code> |172 | <code>throwing event (e.g. javelin, discus, shot put)</code> | <code>extinctive prescription</code> | <code>False</code> |173* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:174 ```json175 {176 "scale": 20.0,177 "similarity_fct": "pairwise_cos_sim"178 }179 ```180 181### Training Hyperparameters182#### Non-Default Hyperparameters183 184- `eval_strategy`: steps185- `per_device_train_batch_size`: 128186- `per_device_eval_batch_size`: 32187- `weight_decay`: 0.01188- `num_train_epochs`: 8189- `warmup_ratio`: 0.1190 191#### All Hyperparameters192<details><summary>Click to expand</summary>193 194- `overwrite_output_dir`: False195- `do_predict`: False196- `eval_strategy`: steps197- `prediction_loss_only`: True198- `per_device_train_batch_size`: 128199- `per_device_eval_batch_size`: 32200- `per_gpu_train_batch_size`: None201- `per_gpu_eval_batch_size`: None202- `gradient_accumulation_steps`: 1203- `eval_accumulation_steps`: None204- `torch_empty_cache_steps`: None205- `learning_rate`: 5e-05206- `weight_decay`: 0.01207- `adam_beta1`: 0.9208- `adam_beta2`: 0.999209- `adam_epsilon`: 1e-08210- `max_grad_norm`: 1.0211- `num_train_epochs`: 8212- `max_steps`: -1213- `lr_scheduler_type`: linear214- `lr_scheduler_kwargs`: {}215- `warmup_ratio`: 0.1216- `warmup_steps`: 0217- `log_level`: passive218- `log_level_replica`: warning219- `log_on_each_node`: True220- `logging_nan_inf_filter`: True221- `save_safetensors`: True222- `save_on_each_node`: False223- `save_only_model`: False224- `restore_callback_states_from_checkpoint`: False225- `no_cuda`: False226- `use_cpu`: False227- `use_mps_device`: False228- `seed`: 42229- `data_seed`: None230- `jit_mode_eval`: False231- `use_ipex`: False232- `bf16`: False233- `fp16`: False234- `fp16_opt_level`: O1235- `half_precision_backend`: auto236- `bf16_full_eval`: False237- `fp16_full_eval`: False238- `tf32`: None239- `local_rank`: 0240- `ddp_backend`: None241- `tpu_num_cores`: None242- `tpu_metrics_debug`: False243- `debug`: []244- `dataloader_drop_last`: False245- `dataloader_num_workers`: 0246- `dataloader_prefetch_factor`: None247- `past_index`: -1248- `disable_tqdm`: False249- `remove_unused_columns`: True250- `label_names`: None251- `load_best_model_at_end`: False252- `ignore_data_skip`: False253- `fsdp`: []254- `fsdp_min_num_params`: 0255- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}256- `fsdp_transformer_layer_cls_to_wrap`: None257- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}258- `deepspeed`: None259- `label_smoothing_factor`: 0.0260- `optim`: adamw_torch261- `optim_args`: None262- `adafactor`: False263- `group_by_length`: False264- `length_column_name`: length265- `ddp_find_unused_parameters`: None266- `ddp_bucket_cap_mb`: None267- `ddp_broadcast_buffers`: False268- `dataloader_pin_memory`: True269- `dataloader_persistent_workers`: False270- `skip_memory_metrics`: True271- `use_legacy_prediction_loop`: False272- `push_to_hub`: False273- `resume_from_checkpoint`: None274- `hub_model_id`: None275- `hub_strategy`: every_save276- `hub_private_repo`: None277- `hub_always_push`: False278- `gradient_checkpointing`: False279- `gradient_checkpointing_kwargs`: None280- `include_inputs_for_metrics`: False281- `include_for_metrics`: []282- `eval_do_concat_batches`: True283- `fp16_backend`: auto284- `push_to_hub_model_id`: None285- `push_to_hub_organization`: None286- `mp_parameters`: 287- `auto_find_batch_size`: False288- `full_determinism`: False289- `torchdynamo`: None290- `ray_scope`: last291- `ddp_timeout`: 1800292- `torch_compile`: False293- `torch_compile_backend`: None294- `torch_compile_mode`: None295- `dispatch_batches`: None296- `split_batches`: None297- `include_tokens_per_second`: False298- `include_num_input_tokens_seen`: False299- `neftune_noise_alpha`: None300- `optim_target_modules`: None301- `batch_eval_metrics`: False302- `eval_on_start`: False303- `use_liger_kernel`: False304- `eval_use_gather_object`: False305- `average_tokens_across_devices`: False306- `prompts`: None307- `batch_sampler`: batch_sampler308- `multi_dataset_batch_sampler`: proportional309 310</details>311 312### Training Logs313<details><summary>Click to expand</summary>314 315| Epoch | Step | Training Loss | Validation Loss | dev_cosine_ap |316|:------:|:------:|:-------------:|:---------------:|:-------------:|317| -1 | -1 | - | - | 0.8061 |318| 0.0145 | 500 | 7.2395 | - | - |319| 0.0291 | 1000 | 7.2421 | - | - |320| 0.0436 | 1500 | 6.5757 | - | - |321| 0.0581 | 2000 | 5.96 | - | - |322| 0.0726 | 2500 | 5.5217 | - | - |323| 0.0872 | 3000 | 5.3224 | - | - |324| 0.1017 | 3500 | 5.2104 | - | - |325| 0.1162 | 4000 | 5.0525 | - | - |326| 0.1308 | 4500 | 5.1228 | - | - |327| 0.1453 | 5000 | 5.0317 | 1.5742 | 0.8818 |328| 0.1598 | 5500 | 4.9875 | - | - |329| 0.1744 | 6000 | 4.85 | - | - |330| 0.1889 | 6500 | 4.9348 | - | - |331| 0.2034 | 7000 | 4.7928 | - | - |332| 0.2179 | 7500 | 4.8412 | - | - |333| 0.2325 | 8000 | 4.8304 | - | - |334| 0.2470 | 8500 | 4.8031 | - | - |335| 0.2615 | 9000 | 4.7567 | - | - |336| 0.2761 | 9500 | 4.7847 | - | - |337| 0.2906 | 10000 | 4.7743 | 1.3281 | 0.9066 |338| 0.3051 | 10500 | 4.6624 | - | - |339| 0.3196 | 11000 | 4.6653 | - | - |340| 0.3342 | 11500 | 4.6047 | - | - |341| 0.3487 | 12000 | 4.5972 | - | - |342| 0.3632 | 12500 | 4.6678 | - | - |343| 0.3778 | 13000 | 4.5873 | - | - |344| 0.3923 | 13500 | 4.6007 | - | - |345| 0.4068 | 14000 | 4.526 | - | - |346| 0.4214 | 14500 | 4.576 | - | - |347| 0.4359 | 15000 | 4.5587 | 1.1674 | 0.9213 |348| 0.4504 | 15500 | 4.4398 | - | - |349| 0.4649 | 16000 | 4.529 | - | - |350| 0.4795 | 16500 | 4.4231 | - | - |351| 0.4940 | 17000 | 4.5204 | - | - |352| 0.5085 | 17500 | 4.508 | - | - |353| 0.5231 | 18000 | 4.4563 | - | - |354| 0.5376 | 18500 | 4.4922 | - | - |355| 0.5521 | 19000 | 4.3455 | - | - |356| 0.5666 | 19500 | 4.393 | - | - |357| 0.5812 | 20000 | 4.3754 | 1.1346 | 0.9267 |358| 0.5957 | 20500 | 4.3033 | - | - |359| 0.6102 | 21000 | 4.4046 | - | - |360| 0.6248 | 21500 | 4.4623 | - | - |361| 0.6393 | 22000 | 4.3426 | - | - |362| 0.6538 | 22500 | 4.3791 | - | - |363| 0.6684 | 23000 | 4.4055 | - | - |364| 0.6829 | 23500 | 4.3898 | - | - |365| 0.6974 | 24000 | 4.3318 | - | - |366| 0.7119 | 24500 | 4.3469 | - | - |367| 0.7265 | 25000 | 4.39 | 1.1003 | 0.9304 |368| 0.7410 | 25500 | 4.2806 | - | - |369| 0.7555 | 26000 | 4.3901 | - | - |370| 0.7701 | 26500 | 4.3526 | - | - |371| 0.7846 | 27000 | 4.2083 | - | - |372| 0.7991 | 27500 | 4.4242 | - | - |373| 0.8136 | 28000 | 4.3139 | - | - |374| 0.8282 | 28500 | 4.2971 | - | - |375| 0.8427 | 29000 | 4.2024 | - | - |376| 0.8572 | 29500 | 4.2684 | - | - |377| 0.8718 | 30000 | 4.3175 | 0.9830 | 0.9365 |378| 0.8863 | 30500 | 4.2168 | - | - |379| 0.9008 | 31000 | 4.1969 | - | - |380| 0.9154 | 31500 | 4.248 | - | - |381| 0.9299 | 32000 | 4.1886 | - | - |382| 0.9444 | 32500 | 4.269 | - | - |383| 0.9589 | 33000 | 4.1733 | - | - |384| 0.9735 | 33500 | 4.1176 | - | - |385| 0.9880 | 34000 | 4.2357 | - | - |386| 1.0025 | 34500 | 4.0826 | - | - |387| 1.0171 | 35000 | 3.6937 | 0.9222 | 0.9416 |388| 1.0316 | 35500 | 3.9462 | - | - |389| 1.0461 | 36000 | 3.8201 | - | - |390| 1.0606 | 36500 | 3.8564 | - | - |391| 1.0752 | 37000 | 3.8252 | - | - |392| 1.0897 | 37500 | 3.8981 | - | - |393| 1.1042 | 38000 | 3.8162 | - | - |394| 1.1188 | 38500 | 3.742 | - | - |395| 1.1333 | 39000 | 3.7388 | - | - |396| 1.1478 | 39500 | 3.852 | - | - |397| 1.1624 | 40000 | 3.7787 | 0.8873 | 0.9440 |398| 1.1769 | 40500 | 3.6863 | - | - |399| 1.1914 | 41000 | 3.7342 | - | - |400| 1.2059 | 41500 | 3.7647 | - | - |401| 1.2205 | 42000 | 3.7589 | - | - |402| 1.2350 | 42500 | 3.7183 | - | - |403| 1.2495 | 43000 | 3.8539 | - | - |404| 1.2641 | 43500 | 3.7406 | - | - |405| 1.2786 | 44000 | 3.7291 | - | - |406| 1.2931 | 44500 | 3.729 | - | - |407| 1.3076 | 45000 | 3.6944 | 0.8696 | 0.9457 |408| 1.3222 | 45500 | 3.8864 | - | - |409| 1.3367 | 46000 | 3.7167 | - | - |410| 1.3512 | 46500 | 3.7737 | - | - |411| 1.3658 | 47000 | 3.7781 | - | - |412| 1.3803 | 47500 | 3.7873 | - | - |413| 1.3948 | 48000 | 3.6664 | - | - |414| 1.4094 | 48500 | 3.8184 | - | - |415| 1.4239 | 49000 | 3.6521 | - | - |416| 1.4384 | 49500 | 3.7833 | - | - |417| 1.4529 | 50000 | 3.7294 | 0.8075 | 0.9504 |418| 1.4675 | 50500 | 3.7328 | - | - |419| 1.4820 | 51000 | 3.7784 | - | - |420| 1.4965 | 51500 | 3.6691 | - | - |421| 1.5111 | 52000 | 3.6275 | - | - |422| 1.5256 | 52500 | 3.7145 | - | - |423| 1.5401 | 53000 | 3.6423 | - | - |424| 1.5546 | 53500 | 3.6464 | - | - |425| 1.5692 | 54000 | 3.6415 | - | - |426| 1.5837 | 54500 | 3.7093 | - | - |427| 1.5982 | 55000 | 3.6996 | 0.7741 | 0.9527 |428| 1.6128 | 55500 | 3.6644 | - | - |429| 1.6273 | 56000 | 3.6496 | - | - |430| 1.6418 | 56500 | 3.6891 | - | - |431| 1.6564 | 57000 | 3.7227 | - | - |432| 1.6709 | 57500 | 3.6413 | - | - |433| 1.6854 | 58000 | 3.6085 | - | - |434| 1.6999 | 58500 | 3.4957 | - | - |435| 1.7145 | 59000 | 3.5888 | - | - |436| 1.7290 | 59500 | 3.6562 | - | - |437| 1.7435 | 60000 | 3.6091 | 0.7441 | 0.9549 |438| 1.7581 | 60500 | 3.4945 | - | - |439| 1.7726 | 61000 | 3.5744 | - | - |440| 1.7871 | 61500 | 3.6632 | - | - |441| 1.8016 | 62000 | 3.5322 | - | - |442| 1.8162 | 62500 | 3.4866 | - | - |443| 1.8307 | 63000 | 3.5391 | - | - |444| 1.8452 | 63500 | 3.4714 | - | - |445| 1.8598 | 64000 | 3.4245 | - | - |446| 1.8743 | 64500 | 3.4765 | - | - |447| 1.8888 | 65000 | 3.4499 | 0.7203 | 0.9563 |448| 1.9034 | 65500 | 3.5459 | - | - |449| 1.9179 | 66000 | 3.6055 | - | - |450| 1.9324 | 66500 | 3.5734 | - | - |451| 1.9469 | 67000 | 3.5724 | - | - |452| 1.9615 | 67500 | 3.5344 | - | - |453| 1.9760 | 68000 | 3.4783 | - | - |454| 1.9905 | 68500 | 3.5332 | - | - |455| 2.0051 | 69000 | 3.1724 | - | - |456| 2.0196 | 69500 | 2.8641 | - | - |457| 2.0341 | 70000 | 2.7543 | 0.7252 | 0.9577 |458| 2.0486 | 70500 | 2.8778 | - | - |459| 2.0632 | 71000 | 2.5721 | - | - |460| 2.0777 | 71500 | 2.7482 | - | - |461| 2.0922 | 72000 | 2.8025 | - | - |462| 2.1068 | 72500 | 2.8993 | - | - |463| 2.1213 | 73000 | 2.9477 | - | - |464| 2.1358 | 73500 | 2.8873 | - | - |465| 2.1504 | 74000 | 2.9593 | - | - |466| 2.1649 | 74500 | 2.8642 | - | - |467| 2.1794 | 75000 | 2.9113 | 0.7252 | 0.9582 |468| 2.1939 | 75500 | 2.8282 | - | - |469| 2.2085 | 76000 | 2.9086 | - | - |470| 2.2230 | 76500 | 2.7911 | - | - |471| 2.2375 | 77000 | 2.9013 | - | - |472| 2.2521 | 77500 | 2.9883 | - | - |473| 2.2666 | 78000 | 2.7996 | - | - |474| 2.2811 | 78500 | 2.9005 | - | - |475| 2.2956 | 79000 | 2.8725 | - | - |476| 2.3102 | 79500 | 2.9003 | - | - |477| 2.3247 | 80000 | 3.0029 | 0.6799 | 0.9607 |478| 2.3392 | 80500 | 2.9904 | - | - |479| 2.3538 | 81000 | 2.9155 | - | - |480| 2.3683 | 81500 | 2.933 | - | - |481| 2.3828 | 82000 | 2.8691 | - | - |482| 2.3973 | 82500 | 3.003 | - | - |483| 2.4119 | 83000 | 2.9573 | - | - |484| 2.4264 | 83500 | 2.8678 | - | - |485| 2.4409 | 84000 | 3.0882 | - | - |486| 2.4555 | 84500 | 2.8722 | - | - |487| 2.4700 | 85000 | 2.9527 | 0.6760 | 0.9610 |488| 2.4845 | 85500 | 3.1515 | - | - |489| 2.4991 | 86000 | 2.9227 | - | - |490| 2.5136 | 86500 | 2.9474 | - | - |491| 2.5281 | 87000 | 2.9981 | - | - |492| 2.5426 | 87500 | 2.8989 | - | - |493| 2.5572 | 88000 | 2.8141 | - | - |494| 2.5717 | 88500 | 3.0488 | - | - |495| 2.5862 | 89000 | 2.8426 | - | - |496| 2.6008 | 89500 | 2.7394 | - | - |497| 2.6153 | 90000 | 3.0399 | 0.6430 | 0.9628 |498| 2.6298 | 90500 | 2.9426 | - | - |499| 2.6443 | 91000 | 2.7746 | - | - |500| 2.6589 | 91500 | 2.9781 | - | - |501| 2.6734 | 92000 | 2.8177 | - | - |502| 2.6879 | 92500 | 2.6764 | - | - |503| 2.7025 | 93000 | 2.8852 | - | - |504| 2.7170 | 93500 | 2.8658 | - | - |505| 2.7315 | 94000 | 2.9031 | - | - |506| 2.7461 | 94500 | 2.9051 | - | - |507| 2.7606 | 95000 | 2.9715 | 0.6347 | 0.9636 |508| 2.7751 | 95500 | 2.8294 | - | - |509| 2.7896 | 96000 | 2.9833 | - | - |510| 2.8042 | 96500 | 2.8931 | - | - |511| 2.8187 | 97000 | 2.866 | - | - |512| 2.8332 | 97500 | 2.7796 | - | - |513| 2.8478 | 98000 | 2.7783 | - | - |514| 2.8623 | 98500 | 2.9983 | - | - |515| 2.8768 | 99000 | 2.965 | - | - |516| 2.8913 | 99500 | 2.9125 | - | - |517| 2.9059 | 100000 | 2.8308 | 0.6162 | 0.9649 |518| 2.9204 | 100500 | 2.7666 | - | - |519| 2.9349 | 101000 | 2.8829 | - | - |520| 2.9495 | 101500 | 2.7808 | - | - |521| 2.9640 | 102000 | 3.0559 | - | - |522| 2.9785 | 102500 | 2.8531 | - | - |523| 2.9931 | 103000 | 2.8534 | - | - |524| 3.0076 | 103500 | 2.3948 | - | - |525| 3.0221 | 104000 | 1.9878 | - | - |526| 3.0366 | 104500 | 2.204 | - | - |527| 3.0512 | 105000 | 2.0951 | 0.6358 | 0.9651 |528| 3.0657 | 105500 | 2.1723 | - | - |529| 3.0802 | 106000 | 2.096 | - | - |530| 3.0948 | 106500 | 2.1398 | - | - |531| 3.1093 | 107000 | 2.1534 | - | - |532| 3.1238 | 107500 | 2.0605 | - | - |533| 3.1383 | 108000 | 1.9515 | - | - |534| 3.1529 | 108500 | 2.1798 | - | - |535| 3.1674 | 109000 | 2.1395 | - | - |536| 3.1819 | 109500 | 2.0357 | - | - |537| 3.1965 | 110000 | 2.0579 | 0.6275 | 0.9656 |538| 3.2110 | 110500 | 2.2834 | - | - |539| 3.2255 | 111000 | 2.1215 | - | - |540| 3.2401 | 111500 | 2.3135 | - | - |541| 3.2546 | 112000 | 2.1642 | - | - |542| 3.2691 | 112500 | 2.1095 | - | - |543| 3.2836 | 113000 | 2.1022 | - | - |544| 3.2982 | 113500 | 2.2954 | - | - |545| 3.3127 | 114000 | 2.2834 | - | - |546| 3.3272 | 114500 | 2.2489 | - | - |547| 3.3418 | 115000 | 2.2317 | 0.6205 | 0.9663 |548| 3.3563 | 115500 | 2.234 | - | - |549| 3.3708 | 116000 | 2.1769 | - | - |550| 3.3853 | 116500 | 2.1369 | - | - |551| 3.3999 | 117000 | 2.1962 | - | - |552| 3.4144 | 117500 | 2.1586 | - | - |553| 3.4289 | 118000 | 2.2802 | - | - |554| 3.4435 | 118500 | 2.2446 | - | - |555| 3.4580 | 119000 | 2.3673 | - | - |556| 3.4725 | 119500 | 2.1549 | - | - |557| 3.4871 | 120000 | 2.2963 | 0.5948 | 0.9672 |558| 3.5016 | 120500 | 2.331 | - | - |559| 3.5161 | 121000 | 2.2441 | - | - |560| 3.5306 | 121500 | 2.0613 | - | - |561| 3.5452 | 122000 | 2.2732 | - | - |562| 3.5597 | 122500 | 2.1462 | - | - |563| 3.5742 | 123000 | 2.2862 | - | - |564| 3.5888 | 123500 | 2.466 | - | - |565| 3.6033 | 124000 | 2.1136 | - | - |566| 3.6178 | 124500 | 2.2851 | - | - |567| 3.6323 | 125000 | 2.2898 | 0.5887 | 0.9677 |568| 3.6469 | 125500 | 2.1318 | - | - |569| 3.6614 | 126000 | 2.2125 | - | - |570| 3.6759 | 126500 | 2.2985 | - | - |571| 3.6905 | 127000 | 2.2355 | - | - |572| 3.7050 | 127500 | 2.1965 | - | - |573| 3.7195 | 128000 | 2.2711 | - | - |574| 3.7341 | 128500 | 2.2094 | - | - |575| 3.7486 | 129000 | 2.1588 | - | - |576| 3.7631 | 129500 | 2.3413 | - | - |577| 3.7776 | 130000 | 2.1223 | 0.5878 | 0.9683 |578| 3.7922 | 130500 | 2.1582 | - | - |579| 3.8067 | 131000 | 2.3648 | - | - |580| 3.8212 | 131500 | 2.2182 | - | - |581| 3.8358 | 132000 | 2.1239 | - | - |582| 3.8503 | 132500 | 2.0056 | - | - |583| 3.8648 | 133000 | 2.1289 | - | - |584| 3.8793 | 133500 | 2.223 | - | - |585| 3.8939 | 134000 | 2.3067 | - | - |586| 3.9084 | 134500 | 2.2172 | - | - |587| 3.9229 | 135000 | 2.2992 | 0.5534 | 0.9699 |588| 3.9375 | 135500 | 2.1945 | - | - |589| 3.9520 | 136000 | 2.2532 | - | - |590| 3.9665 | 136500 | 2.3272 | - | - |591| 3.9811 | 137000 | 2.2678 | - | - |592| 3.9956 | 137500 | 2.2451 | - | - |593| 4.0101 | 138000 | 1.506 | - | - |594| 4.0246 | 138500 | 1.552 | - | - |595| 4.0392 | 139000 | 1.5056 | - | - |596| 4.0537 | 139500 | 1.5867 | - | - |597| 4.0682 | 140000 | 1.4977 | 0.5668 | 0.9697 |598| 4.0828 | 140500 | 1.5145 | - | - |599| 4.0973 | 141000 | 1.571 | - | - |600| 4.1118 | 141500 | 1.5091 | - | - |601| 4.1263 | 142000 | 1.5696 | - | - |602| 4.1409 | 142500 | 1.6053 | - | - |603| 4.1554 | 143000 | 1.5816 | - | - |604| 4.1699 | 143500 | 1.6723 | - | - |605| 4.1845 | 144000 | 1.5638 | - | - |606| 4.1990 | 144500 | 1.5457 | - | - |607| 4.2135 | 145000 | 1.5442 | 0.5663 | 0.9698 |608| 4.2281 | 145500 | 1.6303 | - | - |609| 4.2426 | 146000 | 1.4715 | - | - |610| 4.2571 | 146500 | 1.5385 | - | - |611| 4.2716 | 147000 | 1.6144 | - | - |612| 4.2862 | 147500 | 1.4881 | - | - |613| 4.3007 | 148000 | 1.8148 | - | - |614| 4.3152 | 148500 | 1.5511 | - | - |615| 4.3298 | 149000 | 1.6536 | - | - |616| 4.3443 | 149500 | 1.5755 | - | - |617| 4.3588 | 150000 | 1.6997 | 0.5608 | 0.9702 |618| 4.3733 | 150500 | 1.6931 | - | - |619| 4.3879 | 151000 | 1.5777 | - | - |620| 4.4024 | 151500 | 1.7588 | - | - |621| 4.4169 | 152000 | 1.5043 | - | - |622| 4.4315 | 152500 | 1.5527 | - | - |623| 4.4460 | 153000 | 1.5128 | - | - |624| 4.4605 | 153500 | 1.5893 | - | - |625| 4.4751 | 154000 | 1.6465 | - | - |626| 4.4896 | 154500 | 1.6211 | - | - |627| 4.5041 | 155000 | 1.5675 | 0.5623 | 0.9704 |628| 4.5186 | 155500 | 1.752 | - | - |629| 4.5332 | 156000 | 1.8182 | - | - |630| 4.5477 | 156500 | 1.5368 | - | - |631| 4.5622 | 157000 | 1.6635 | - | - |632| 4.5768 | 157500 | 1.5425 | - | - |633| 4.5913 | 158000 | 1.5988 | - | - |634| 4.6058 | 158500 | 1.7011 | - | - |635| 4.6203 | 159000 | 1.5353 | - | - |636| 4.6349 | 159500 | 1.625 | - | - |637| 4.6494 | 160000 | 1.5483 | 0.5426 | 0.9714 |638| 4.6639 | 160500 | 1.6127 | - | - |639| 4.6785 | 161000 | 1.6512 | - | - |640| 4.6930 | 161500 | 1.7213 | - | - |641| 4.7075 | 162000 | 1.5976 | - | - |642| 4.7221 | 162500 | 1.5711 | - | - |643| 4.7366 | 163000 | 1.5911 | - | - |644| 4.7511 | 163500 | 1.6364 | - | - |645| 4.7656 | 164000 | 1.6361 | - | - |646| 4.7802 | 164500 | 1.7027 | - | - |647| 4.7947 | 165000 | 1.6462 | 0.5388 | 0.9717 |648| 4.8092 | 165500 | 1.7102 | - | - |649| 4.8238 | 166000 | 1.6149 | - | - |650| 4.8383 | 166500 | 1.5491 | - | - |651| 4.8528 | 167000 | 1.6389 | - | - |652| 4.8673 | 167500 | 1.5092 | - | - |653| 4.8819 | 168000 | 1.6771 | - | - |654| 4.8964 | 168500 | 1.6812 | - | - |655| 4.9109 | 169000 | 1.6414 | - | - |656| 4.9255 | 169500 | 1.6066 | - | - |657| 4.9400 | 170000 | 1.4729 | 0.5236 | 0.9724 |658| 4.9545 | 170500 | 1.6032 | - | - |659| 4.9691 | 171000 | 1.6274 | - | - |660| 4.9836 | 171500 | 1.8478 | - | - |661| 4.9981 | 172000 | 1.6356 | - | - |662| 5.0126 | 172500 | 1.1942 | - | - |663| 5.0272 | 173000 | 1.1838 | - | - |664| 5.0417 | 173500 | 1.0514 | - | - |665| 5.0562 | 174000 | 1.0647 | - | - |666| 5.0708 | 174500 | 1.0718 | - | - |667| 5.0853 | 175000 | 1.0162 | 0.5385 | 0.9720 |668| 5.0998 | 175500 | 1.0253 | - | - |669| 5.1143 | 176000 | 1.115 | - | - |670| 5.1289 | 176500 | 1.0504 | - | - |671| 5.1434 | 177000 | 1.1573 | - | - |672| 5.1579 | 177500 | 1.0937 | - | - |673| 5.1725 | 178000 | 1.0939 | - | - |674| 5.1870 | 178500 | 1.0392 | - | - |675| 5.2015 | 179000 | 1.0852 | - | - |676| 5.2161 | 179500 | 1.165 | - | - |677| 5.2306 | 180000 | 1.1048 | 0.5291 | 0.9723 |678| 5.2451 | 180500 | 1.1814 | - | - |679| 5.2596 | 181000 | 1.2639 | - | - |680| 5.2742 | 181500 | 1.1395 | - | - |681| 5.2887 | 182000 | 1.1452 | - | - |682| 5.3032 | 182500 | 1.2131 | - | - |683| 5.3178 | 183000 | 1.236 | - | - |684| 5.3323 | 183500 | 1.1449 | - | - |685| 5.3468 | 184000 | 1.1425 | - | - |686| 5.3613 | 184500 | 1.2328 | - | - |687| 5.3759 | 185000 | 1.1114 | 0.5252 | 0.9727 |688 689</details>690 691### Framework Versions692- Python: 3.9.21693- Sentence Transformers: 3.4.0694- Transformers: 4.48.1695- PyTorch: 2.5.1696- Accelerate: 1.3.0697- Datasets: 3.2.0698- Tokenizers: 0.21.0