jansowa/albert-base-v2-multilingual-en-pl

SentenceTransformer based on albert/albert-base-v2

This is a sentence-transformers model finetuned from albert/albert-base-v2 on the en-pl 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: albert/albert-base-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • en-pl
• Languages: en, multilingual, ar, bg, ca, cs, da, de, el, es, et, fa, fi, fr, gl, gu, he, hi, hr, hu, hy, id, it, ja, ka, ko, ku, lt, lv, mk, mn, mr, ms, my, nb, nl, pl, pt, ro, ru, sk, sl, sq, sr, sv, th, tr, uk, ur, vi, zh

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False, 'architecture': 'AlbertModel'})
  (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:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("jansowa/albert-base-v2-multilingual-en-pl")
# Run inference
sentences = [
    'This is a diagram of the U.S. counterinsurgency strategy in Afghanistan.',
    'To jest diagram przeciwpartyzanckiej strategii w Afganistanie.',
    'Biedni ludzie, ludzie, których prawa człowieka zostały naruszone, brzemię tego to strata godności. Brak godności.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.9598, 0.5281],
#         [0.9598, 1.0000, 0.5946],
#         [0.5281, 0.5946, 1.0000]])

Evaluation

Metrics

Knowledge Distillation

• Dataset: en-pl
• Evaluated with MSEEvaluator

Metric	Value
negative_mse	-37.0147

Translation

• Dataset: en-pl
• Evaluated with TranslationEvaluator

Metric	Value
src2trg_accuracy	0.7046
trg2src_accuracy	0.6613
mean_accuracy	0.683

Training Details

Training Dataset

en-pl

• Dataset: en-pl at 0c70bc6
• Size: 292,290 training samples
• Columns: english, non_english, and label
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	min: 4 tokens mean: 24.47 tokens max: 256 tokens	min: 5 tokens mean: 40.48 tokens max: 256 tokens	size: 768 elements

• Samples:

  english	non_english	label
  And then there are certain conceptual things that can also benefit from hand calculating, but I think they're relatively small in number.	Są też pewne zadania koncepcyjne, w których ręczne kalkulacje mogą być przydatne, ale jest ich stosunkowo niewiele.	[0.1842687577009201, -0.27380749583244324, 1.380724310874939, 0.5485912561416626, -0.5771370530128479, ...]
  One thing I often ask about is ancient Greek and how this relates.	Często zadaję pytania o starożytną Grekę i co do tego ma.	[-0.22509485483169556, -0.8029794096946716, -0.26132631301879883, -0.1386972814798355, -0.4966896176338196, ...]
  See, the thing we're doing right now is we're forcing people to learn mathematics.	Zwróćcie uwagę, to co teraz robimy jest zmuszaniem ludzi do nauki matematyki.	[0.701093316078186, -0.31419914960861206, -0.8785677552223206, -0.3886241614818573, 0.7142088413238525, ...]

• Loss: MSELoss

Evaluation Dataset

en-pl

• Dataset: en-pl at 0c70bc6
• Size: 992 evaluation samples
• Columns: english, non_english, and label
• Approximate statistics based on the first 992 samples:

  	english	non_english	label
  type	string	string	list
  details	min: 4 tokens mean: 25.03 tokens max: 253 tokens	min: 5 tokens mean: 45.15 tokens max: 256 tokens	size: 768 elements

• Samples:

  english	non_english	label
  Thank you so much, Chris.	Bardzo dziękuję, Chris.	[0.26128441095352173, -0.8462327122688293, -0.4199201762676239, 0.5228638648986816, 1.1514642238616943, ...]
  And it's truly a great honor to have the opportunity to come to this stage twice; I'm extremely grateful.	To prawdziwy zaszczyt mieć możliwość drugi raz stanąć w tym miejscu. Jestem niezwykle wdzięczny.	[0.22651851177215576, 0.283356636762619, -1.0012000799179077, -0.013265828602015972, 0.08300188928842545, ...]
  I have been blown away by this conference, and I want to thank all of you for the many nice comments about what I had to say the other night.	Jestem zachwycony tą konferencją. Chce Wam wszystkim podziękować za miłe komentarze dotyczące mojej wypowiedzi poprzedniego wieczoru.	[0.17416058480739594, -0.40953880548477173, -0.62795090675354, 0.35556134581565857, 0.40010693669319153, ...]

• Loss: MSELoss

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• fp16: True

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 8
• per_device_eval_batch_size: 8
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 2e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 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
• 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: 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}
• 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
• 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
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	en-pl loss	en-pl_negative_mse	en-pl_mean_accuracy
0.0027	100	1.1564	-	-	-
0.0055	200	0.8704	-	-	-
0.0082	300	0.6851	-	-	-
0.0109	400	0.5806	-	-	-
0.0137	500	0.5265	-	-	-
0.0164	600	0.4866	-	-	-
0.0192	700	0.4744	-	-	-
0.0219	800	0.4665	-	-	-
0.0246	900	0.4637	-	-	-
0.0274	1000	0.4616	-	-	-
0.0301	1100	0.461	-	-	-
0.0328	1200	0.4624	-	-	-
0.0356	1300	0.4605	-	-	-
0.0383	1400	0.4596	-	-	-
0.0411	1500	0.459	-	-	-
0.0438	1600	0.455	-	-	-
0.0465	1700	0.4567	-	-	-
0.0493	1800	0.4567	-	-	-
0.0520	1900	0.4541	-	-	-
0.0547	2000	0.4559	-	-	-
0.0575	2100	0.456	-	-	-
0.0602	2200	0.4535	-	-	-
0.0629	2300	0.4495	-	-	-
0.0657	2400	0.4515	-	-	-
0.0684	2500	0.4472	-	-	-
0.0712	2600	0.4486	-	-	-
0.0739	2700	0.4468	-	-	-
0.0766	2800	0.4432	-	-	-
0.0794	2900	0.4443	-	-	-
0.0821	3000	0.4437	-	-	-
0.0848	3100	0.439	-	-	-
0.0876	3200	0.4379	-	-	-
0.0903	3300	0.4345	-	-	-
0.0931	3400	0.4335	-	-	-
0.0958	3500	0.4322	-	-	-
0.0985	3600	0.4326	-	-	-
0.1013	3700	0.4319	-	-	-
0.1040	3800	0.4298	-	-	-
0.1067	3900	0.4243	-	-	-
0.1095	4000	0.4243	-	-	-
0.1122	4100	0.4226	-	-	-
0.1150	4200	0.4216	-	-	-
0.1177	4300	0.4208	-	-	-
0.1204	4400	0.4245	-	-	-
0.1232	4500	0.4236	-	-	-
0.1259	4600	0.4176	-	-	-
0.1286	4700	0.4175	-	-	-
0.1314	4800	0.4148	-	-	-
0.1341	4900	0.4163	-	-	-
0.1368	5000	0.4127	0.4204	-44.0072	0.1240
0.1396	5100	0.4089	-	-	-
0.1423	5200	0.4145	-	-	-
0.1451	5300	0.4084	-	-	-
0.1478	5400	0.4082	-	-	-
0.1505	5500	0.4062	-	-	-
0.1533	5600	0.4056	-	-	-
0.1560	5700	0.4039	-	-	-
0.1587	5800	0.4059	-	-	-
0.1615	5900	0.405	-	-	-
0.1642	6000	0.4003	-	-	-
0.1670	6100	0.4003	-	-	-
0.1697	6200	0.3994	-	-	-
0.1724	6300	0.4	-	-	-
0.1752	6400	0.3965	-	-	-
0.1779	6500	0.3964	-	-	-
0.1806	6600	0.3948	-	-	-
0.1834	6700	0.3955	-	-	-
0.1861	6800	0.3959	-	-	-
0.1888	6900	0.3942	-	-	-
0.1916	7000	0.3922	-	-	-
0.1943	7100	0.3933	-	-	-
0.1971	7200	0.395	-	-	-
0.1998	7300	0.3932	-	-	-
0.2025	7400	0.3881	-	-	-
0.2053	7500	0.3884	-	-	-
0.2080	7600	0.3862	-	-	-
0.2107	7700	0.3864	-	-	-
0.2135	7800	0.3879	-	-	-
0.2162	7900	0.3895	-	-	-
0.2190	8000	0.3847	-	-	-
0.2217	8100	0.3856	-	-	-
0.2244	8200	0.3863	-	-	-
0.2272	8300	0.3859	-	-	-
0.2299	8400	0.3821	-	-	-
0.2326	8500	0.3825	-	-	-
0.2354	8600	0.3799	-	-	-
0.2381	8700	0.381	-	-	-
0.2409	8800	0.3824	-	-	-
0.2436	8900	0.3811	-	-	-
0.2463	9000	0.3774	-	-	-
0.2491	9100	0.3781	-	-	-
0.2518	9200	0.3792	-	-	-
0.2545	9300	0.3769	-	-	-
0.2573	9400	0.376	-	-	-
0.2600	9500	0.3799	-	-	-
0.2627	9600	0.3737	-	-	-
0.2655	9700	0.3757	-	-	-
0.2682	9800	0.3753	-	-	-
0.2710	9900	0.3761	-	-	-
0.2737	10000	0.3701	0.3808	-41.4255	0.3191
0.2764	10100	0.3718	-	-	-
0.2792	10200	0.3701	-	-	-
0.2819	10300	0.3704	-	-	-
0.2846	10400	0.3724	-	-	-
0.2874	10500	0.3725	-	-	-
0.2901	10600	0.3726	-	-	-
0.2929	10700	0.3679	-	-	-
0.2956	10800	0.3671	-	-	-
0.2983	10900	0.368	-	-	-
0.3011	11000	0.3711	-	-	-
0.3038	11100	0.3696	-	-	-
0.3065	11200	0.3677	-	-	-
0.3093	11300	0.3651	-	-	-
0.3120	11400	0.365	-	-	-
0.3147	11500	0.3635	-	-	-
0.3175	11600	0.3595	-	-	-
0.3202	11700	0.363	-	-	-
0.3230	11800	0.3644	-	-	-
0.3257	11900	0.3649	-	-	-
0.3284	12000	0.3623	-	-	-
0.3312	12100	0.3634	-	-	-
0.3339	12200	0.3616	-	-	-
0.3366	12300	0.3644	-	-	-
0.3394	12400	0.3608	-	-	-
0.3421	12500	0.3601	-	-	-
0.3449	12600	0.3623	-	-	-
0.3476	12700	0.3606	-	-	-
0.3503	12800	0.3585	-	-	-
0.3531	12900	0.3622	-	-	-
0.3558	13000	0.361	-	-	-
0.3585	13100	0.3595	-	-	-
0.3613	13200	0.3569	-	-	-
0.3640	13300	0.3597	-	-	-
0.3668	13400	0.3586	-	-	-
0.3695	13500	0.3577	-	-	-
0.3722	13600	0.3569	-	-	-
0.3750	13700	0.3546	-	-	-
0.3777	13800	0.3546	-	-	-
0.3804	13900	0.3552	-	-	-
0.3832	14000	0.3535	-	-	-
0.3859	14100	0.3566	-	-	-
0.3886	14200	0.3556	-	-	-
0.3914	14300	0.3548	-	-	-
0.3941	14400	0.3529	-	-	-
0.3969	14500	0.3549	-	-	-
0.3996	14600	0.3539	-	-	-
0.4023	14700	0.3508	-	-	-
0.4051	14800	0.3536	-	-	-
0.4078	14900	0.3528	-	-	-
0.4105	15000	0.3548	0.3599	-40.0086	0.4451
0.4133	15100	0.3523	-	-	-
0.4160	15200	0.3483	-	-	-
0.4188	15300	0.3507	-	-	-
0.4215	15400	0.3507	-	-	-
0.4242	15500	0.3516	-	-	-
0.4270	15600	0.3503	-	-	-
0.4297	15700	0.3476	-	-	-
0.4324	15800	0.3484	-	-	-
0.4352	15900	0.3487	-	-	-
0.4379	16000	0.3473	-	-	-
0.4406	16100	0.3501	-	-	-
0.4434	16200	0.3481	-	-	-
0.4461	16300	0.3462	-	-	-
0.4489	16400	0.347	-	-	-
0.4516	16500	0.3458	-	-	-
0.4543	16600	0.3485	-	-	-
0.4571	16700	0.3461	-	-	-
0.4598	16800	0.3483	-	-	-
0.4625	16900	0.3456	-	-	-
0.4653	17000	0.3454	-	-	-
0.4680	17100	0.344	-	-	-
0.4708	17200	0.344	-	-	-
0.4735	17300	0.3417	-	-	-
0.4762	17400	0.3469	-	-	-
0.4790	17500	0.3465	-	-	-
0.4817	17600	0.3438	-	-	-
0.4844	17700	0.3437	-	-	-
0.4872	17800	0.3413	-	-	-
0.4899	17900	0.3425	-	-	-
0.4927	18000	0.3429	-	-	-
0.4954	18100	0.3449	-	-	-
0.4981	18200	0.3425	-	-	-
0.5009	18300	0.3431	-	-	-
0.5036	18400	0.3431	-	-	-
0.5063	18500	0.3429	-	-	-
0.5091	18600	0.343	-	-	-
0.5118	18700	0.3413	-	-	-
0.5145	18800	0.3425	-	-	-
0.5173	18900	0.3386	-	-	-
0.5200	19000	0.3415	-	-	-
0.5228	19100	0.341	-	-	-
0.5255	19200	0.3395	-	-	-
0.5282	19300	0.3413	-	-	-
0.5310	19400	0.3412	-	-	-
0.5337	19500	0.3387	-	-	-
0.5364	19600	0.3413	-	-	-
0.5392	19700	0.3383	-	-	-
0.5419	19800	0.3414	-	-	-
0.5447	19900	0.3377	-	-	-
0.5474	20000	0.341	0.3462	-38.8721	0.5459
0.5501	20100	0.3364	-	-	-
0.5529	20200	0.3377	-	-	-
0.5556	20300	0.3362	-	-	-
0.5583	20400	0.338	-	-	-
0.5611	20500	0.3326	-	-	-
0.5638	20600	0.3362	-	-	-
0.5665	20700	0.3368	-	-	-
0.5693	20800	0.3379	-	-	-
0.5720	20900	0.3362	-	-	-
0.5748	21000	0.334	-	-	-
0.5775	21100	0.3389	-	-	-
0.5802	21200	0.3361	-	-	-
0.5830	21300	0.3358	-	-	-
0.5857	21400	0.3333	-	-	-
0.5884	21500	0.3349	-	-	-
0.5912	21600	0.3332	-	-	-
0.5939	21700	0.3354	-	-	-
0.5967	21800	0.3334	-	-	-
0.5994	21900	0.3324	-	-	-
0.6021	22000	0.3317	-	-	-
0.6049	22100	0.3312	-	-	-
0.6076	22200	0.3352	-	-	-
0.6103	22300	0.333	-	-	-
0.6131	22400	0.3358	-	-	-
0.6158	22500	0.332	-	-	-
0.6186	22600	0.3321	-	-	-
0.6213	22700	0.3327	-	-	-
0.6240	22800	0.3312	-	-	-
0.6268	22900	0.3317	-	-	-
0.6295	23000	0.3277	-	-	-
0.6322	23100	0.3334	-	-	-
0.6350	23200	0.3313	-	-	-
0.6377	23300	0.331	-	-	-
0.6404	23400	0.3326	-	-	-
0.6432	23500	0.3325	-	-	-
0.6459	23600	0.3288	-	-	-
0.6487	23700	0.331	-	-	-
0.6514	23800	0.3315	-	-	-
0.6541	23900	0.3312	-	-	-
0.6569	24000	0.329	-	-	-
0.6596	24100	0.3263	-	-	-
0.6623	24200	0.3326	-	-	-
0.6651	24300	0.3297	-	-	-
0.6678	24400	0.3251	-	-	-
0.6706	24500	0.3309	-	-	-
0.6733	24600	0.3302	-	-	-
0.6760	24700	0.3274	-	-	-
0.6788	24800	0.3278	-	-	-
0.6815	24900	0.3268	-	-	-
0.6842	25000	0.3283	0.3376	-38.1403	0.6129
0.6870	25100	0.3278	-	-	-
0.6897	25200	0.3285	-	-	-
0.6924	25300	0.3288	-	-	-
0.6952	25400	0.3275	-	-	-
0.6979	25500	0.327	-	-	-
0.7007	25600	0.328	-	-	-
0.7034	25700	0.3292	-	-	-
0.7061	25800	0.3255	-	-	-
0.7089	25900	0.3279	-	-	-
0.7116	26000	0.3276	-	-	-
0.7143	26100	0.3254	-	-	-
0.7171	26200	0.3254	-	-	-
0.7198	26300	0.3237	-	-	-
0.7226	26400	0.3261	-	-	-
0.7253	26500	0.3247	-	-	-
0.7280	26600	0.3277	-	-	-
0.7308	26700	0.324	-	-	-
0.7335	26800	0.3262	-	-	-
0.7362	26900	0.3223	-	-	-
0.7390	27000	0.3205	-	-	-
0.7417	27100	0.3265	-	-	-
0.7445	27200	0.3234	-	-	-
0.7472	27300	0.3228	-	-	-
0.7499	27400	0.3202	-	-	-
0.7527	27500	0.3234	-	-	-
0.7554	27600	0.3239	-	-	-
0.7581	27700	0.323	-	-	-
0.7609	27800	0.3232	-	-	-
0.7636	27900	0.324	-	-	-
0.7663	28000	0.3239	-	-	-
0.7691	28100	0.3224	-	-	-
0.7718	28200	0.3258	-	-	-
0.7746	28300	0.3259	-	-	-
0.7773	28400	0.3229	-	-	-
0.7800	28500	0.3266	-	-	-
0.7828	28600	0.3212	-	-	-
0.7855	28700	0.3243	-	-	-
0.7882	28800	0.3237	-	-	-
0.7910	28900	0.3225	-	-	-
0.7937	29000	0.3233	-	-	-
0.7965	29100	0.3249	-	-	-
0.7992	29200	0.3246	-	-	-
0.8019	29300	0.321	-	-	-
0.8047	29400	0.3263	-	-	-
0.8074	29500	0.3244	-	-	-
0.8101	29600	0.3232	-	-	-
0.8129	29700	0.3212	-	-	-
0.8156	29800	0.3235	-	-	-
0.8183	29900	0.3197	-	-	-
0.8211	30000	0.3219	0.3297	-37.3467	0.6714
0.8238	30100	0.3238	-	-	-
0.8266	30200	0.3243	-	-	-
0.8293	30300	0.3238	-	-	-
0.8320	30400	0.3194	-	-	-
0.8348	30500	0.3198	-	-	-
0.8375	30600	0.3227	-	-	-
0.8402	30700	0.3199	-	-	-
0.8430	30800	0.3209	-	-	-
0.8457	30900	0.3212	-	-	-
0.8485	31000	0.3182	-	-	-
0.8512	31100	0.3214	-	-	-
0.8539	31200	0.3203	-	-	-
0.8567	31300	0.3246	-	-	-
0.8594	31400	0.3171	-	-	-
0.8621	31500	0.3208	-	-	-
0.8649	31600	0.3203	-	-	-
0.8676	31700	0.319	-	-	-
0.8704	31800	0.3179	-	-	-
0.8731	31900	0.3187	-	-	-
0.8758	32000	0.3197	-	-	-
0.8786	32100	0.319	-	-	-
0.8813	32200	0.3214	-	-	-
0.8840	32300	0.3205	-	-	-
0.8868	32400	0.3179	-	-	-
0.8895	32500	0.3197	-	-	-
0.8922	32600	0.3197	-	-	-
0.8950	32700	0.3187	-	-	-
0.8977	32800	0.3195	-	-	-
0.9005	32900	0.3193	-	-	-
0.9032	33000	0.3188	-	-	-
0.9059	33100	0.3166	-	-	-
0.9087	33200	0.3186	-	-	-
0.9114	33300	0.3182	-	-	-
0.9141	33400	0.3167	-	-	-
0.9169	33500	0.3203	-	-	-
0.9196	33600	0.3189	-	-	-
0.9224	33700	0.3177	-	-	-
0.9251	33800	0.3174	-	-	-
0.9278	33900	0.3194	-	-	-
0.9306	34000	0.318	-	-	-
0.9333	34100	0.3171	-	-	-
0.9360	34200	0.3185	-	-	-
0.9388	34300	0.3175	-	-	-
0.9415	34400	0.3181	-	-	-
0.9442	34500	0.3219	-	-	-
0.9470	34600	0.3137	-	-	-
0.9497	34700	0.3164	-	-	-
0.9525	34800	0.3161	-	-	-
0.9552	34900	0.3177	-	-	-
0.9579	35000	0.3165	0.3260	-37.0147	0.6830
0.9607	35100	0.3181	-	-	-
0.9634	35200	0.3161	-	-	-
0.9661	35300	0.3156	-	-	-
0.9689	35400	0.3152	-	-	-
0.9716	35500	0.3186	-	-	-
0.9744	35600	0.3197	-	-	-
0.9771	35700	0.3191	-	-	-
0.9798	35800	0.3161	-	-	-
0.9826	35900	0.3184	-	-	-
0.9853	36000	0.3166	-	-	-
0.9880	36100	0.316	-	-	-
0.9908	36200	0.3194	-	-	-
0.9935	36300	0.3158	-	-	-
0.9963	36400	0.3187	-	-	-
0.9990	36500	0.317	-	-	-

Framework Versions

• Python: 3.13.2
• Sentence Transformers: 5.1.0
• Transformers: 4.51.3
• PyTorch: 2.6.0+cu126
• Accelerate: 1.4.0
• Datasets: 3.3.2
• Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers

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

MSELoss

@inproceedings{reimers-2020-multilingual-sentence-bert,
    title = "Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2020",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/2004.09813",
}