meandyou200175/sp_chatbot_query

SentenceTransformer based on vinai/phobert-base-v2

This is a sentence-transformers model finetuned from vinai/phobert-base-v2. 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: vinai/phobert-base-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity

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}) with Transformer model: RobertaModel 
  (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("meandyou200175/sp_chatbot_query")
# Run inference
sentences = [
    'cho tôi máy hút bụi công suất hút trên 8kPa và pin chạy ít nhất 40 phút',
    'Robot hút bụi Xiaomi Vacuum X10, lực hút 20000Pa (20kPa), pin 60 phút, Giá: 11.900.000',
    'Lò nướng Sunhouse SHD4260, dung tích 45L, công suất 1600W, Giá 1.150.000',
]
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

Information Retrieval

• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.2614
cosine_accuracy@2	0.4269
cosine_accuracy@5	0.6739
cosine_accuracy@10	0.8594
cosine_accuracy@100	1.0
cosine_precision@1	0.2614
cosine_precision@2	0.2134
cosine_precision@5	0.1348
cosine_precision@10	0.0859
cosine_precision@100	0.01
cosine_recall@1	0.2614
cosine_recall@2	0.4269
cosine_recall@5	0.6739
cosine_recall@10	0.8594
cosine_recall@100	1.0
cosine_ndcg@10	0.5371
cosine_mrr@1	0.2614
cosine_mrr@2	0.3441
cosine_mrr@5	0.412
cosine_mrr@10	0.4367
cosine_mrr@100	0.4452
cosine_map@100	0.4452

Training Details

Training Dataset

Unnamed Dataset

• Size: 14,462 training samples
• Columns: query and positive
• Approximate statistics based on the first 1000 samples:

  	query	positive
  type	string	string
  details	min: 5 tokens mean: 14.29 tokens max: 27 tokens	min: 17 tokens mean: 35.68 tokens max: 133 tokens

• Samples:

  query	positive
  Có cân điện tử y tế dưới 1.068.000.000 VNĐ không	Omron HN-286 - . Cân điện tử sức khỏe, hiển thị cân nặng & BMI, mặt kính chịu lực, pin AA x 2, thiết kế gọn nhẹ, vận hành êm. Kích thước: 300 x 300 x 25 mm. Trọng lượng: 2 kg. Giá: 890.000 VNĐ
  cần nồi cơm điện công suất trên 700W	Nồi cơm điện Sharp KS-11ETV, Công suất 750W, Dung tích 1.1L, Giá: 1.050.000
  cho tôi màn hình máy tính kích thước tối thiểu 23 inch	Màn hình LG UltraGear 27GN950, 27 inch, 4K UHD, 144Hz, Nano IPS, Giá: 16.800.000

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 1,607 evaluation samples
• Columns: query and positive
• Approximate statistics based on the first 1000 samples:

  	query	positive
  type	string	string
  details	min: 5 tokens mean: 14.33 tokens max: 27 tokens	min: 18 tokens mean: 37.23 tokens max: 139 tokens

• Samples:

  query	positive
  mình cần máy lạnh giá trong khoảng 15 đến 20 triệu, công suất trên 12.000 BTU và tiết kiệm điện 5 sao	Điều hòa Panasonic Inverter 1.5HP, Công suất 12.700 BTU, Công nghệ NanoeX, Giá: 18.200.000
  tôi muốn mua máy chiếu độ sáng trên 3.500 lumen và giá nhỏ hơn 19 triệu	Máy chiếu Epson EB-X51, Độ sáng 3.700 lumen, Độ phân giải XGA, Giá: 14.200.000
  Có bộ đồ trang điểm dưới 1.741.500.000 VNĐ không	Sephora Basics Kit - . Bao gồm 12 màu phấn mắt, 4 màu má hồng, 2 màu son, cọ trang điểm, hộp gọn nhẹ, thích hợp đi du lịch, chất liệu an toàn cho da. Kích thước: 300 x 200 x 50 mm. Trọng lượng: 0.8 kg. Giá: 1.290.000 VNĐ

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 2
• per_device_eval_batch_size: 2
• learning_rate: 2e-05
• num_train_epochs: 6
• warmup_ratio: 0.1
• fp16: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 2
• per_device_eval_batch_size: 2
• 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: 6
• 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}
• 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: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss	cosine_ndcg@10
-1	-1	-	-	0.1506
0.0138	100	0.333	-	-
0.0277	200	0.1879	-	-
0.0415	300	0.0525	-	-
0.0553	400	0.0482	-	-
0.0691	500	0.0271	-	-
0.0830	600	0.0398	-	-
0.0968	700	0.0381	-	-
0.1106	800	0.0401	-	-
0.1245	900	0.0158	-	-
0.1383	1000	0.0251	0.0225	0.3056
0.1521	1100	0.0119	-	-
0.1660	1200	0.0133	-	-
0.1798	1300	0.0278	-	-
0.1936	1400	0.0196	-	-
0.2074	1500	0.0008	-	-
0.2213	1600	0.0405	-	-
0.2351	1700	0.0003	-	-
0.2489	1800	0.0034	-	-
0.2628	1900	0.0271	-	-
0.2766	2000	0.0151	0.0160	0.3355
0.2904	2100	0.0019	-	-
0.3042	2200	0.0181	-	-
0.3181	2300	0.0218	-	-
0.3319	2400	0.0105	-	-
0.3457	2500	0.0551	-	-
0.3596	2600	0.0279	-	-
0.3734	2700	0.0205	-	-
0.3872	2800	0.0018	-	-
0.4011	2900	0.0047	-	-
0.4149	3000	0.018	0.0148	0.3718
0.4287	3100	0.0081	-	-
0.4425	3200	0.0145	-	-
0.4564	3300	0.0258	-	-
0.4702	3400	0.0331	-	-
0.4840	3500	0.0122	-	-
0.4979	3600	0.0179	-	-
0.5117	3700	0.0003	-	-
0.5255	3800	0.0223	-	-
0.5393	3900	0.0126	-	-
0.5532	4000	0.0087	0.0201	0.3609
0.5670	4100	0.0139	-	-
0.5808	4200	0.0189	-	-
0.5947	4300	0.0062	-	-
0.6085	4400	0.0092	-	-
0.6223	4500	0.0192	-	-
0.6361	4600	0.0568	-	-
0.6500	4700	0.0128	-	-
0.6638	4800	0.0312	-	-
0.6776	4900	0.0961	-	-
0.6915	5000	0.0311	0.0093	0.3905
0.7053	5100	0.0176	-	-
0.7191	5200	0.0084	-	-
0.7330	5300	0.0329	-	-
0.7468	5400	0.0015	-	-
0.7606	5500	0.0003	-	-
0.7744	5600	0.0153	-	-
0.7883	5700	0.0077	-	-
0.8021	5800	0.0166	-	-
0.8159	5900	0.0079	-	-
0.8298	6000	0.001	0.0083	0.4171
0.8436	6100	0.0227	-	-
0.8574	6200	0.0591	-	-
0.8712	6300	0.0115	-	-
0.8851	6400	0.0342	-	-
0.8989	6500	0.0199	-	-
0.9127	6600	0.0067	-	-
0.9266	6700	0.0206	-	-
0.9404	6800	0.0092	-	-
0.9542	6900	0.0002	-	-
0.9681	7000	0.0132	0.0113	0.4096
0.9819	7100	0.007	-	-
0.9957	7200	0.0001	-	-
1.0095	7300	0.0219	-	-
1.0234	7400	0.0005	-	-
1.0372	7500	0.0246	-	-
1.0510	7600	0.0117	-	-
1.0649	7700	0.0092	-	-
1.0787	7800	0.0004	-	-
1.0925	7900	0.0352	-	-
1.1063	8000	0.0182	0.0102	0.3950
1.1202	8100	0.0487	-	-
1.1340	8200	0.0391	-	-
1.1478	8300	0.0197	-	-
1.1617	8400	0.0124	-	-
1.1755	8500	0.059	-	-
1.1893	8600	0.0269	-	-
1.2032	8700	0.0004	-	-
1.2170	8800	0.0007	-	-
1.2308	8900	0.0035	-	-
1.2446	9000	0.0056	0.0094	0.4364
1.2585	9100	0.018	-	-
1.2723	9200	0.0159	-	-
1.2861	9300	0.011	-	-
1.3000	9400	0.0222	-	-
1.3138	9500	0.0042	-	-
1.3276	9600	0.0107	-	-
1.3414	9700	0.0271	-	-
1.3553	9800	0.0042	-	-
1.3691	9900	0.0135	-	-
1.3829	10000	0.0099	0.0172	0.4031
1.3968	10100	0.039	-	-
1.4106	10200	0.0573	-	-
1.4244	10300	0.0411	-	-
1.4383	10400	0.0096	-	-
1.4521	10500	0.0207	-	-
1.4659	10600	0.0141	-	-
1.4797	10700	0.0081	-	-
1.4936	10800	0.0	-	-
1.5074	10900	0.0081	-	-
1.5212	11000	0.0166	0.0106	0.4550
1.5351	11100	0.0069	-	-
1.5489	11200	0.0103	-	-
1.5627	11300	0.016	-	-
1.5765	11400	0.0138	-	-
1.5904	11500	0.0023	-	-
1.6042	11600	0.0005	-	-
1.6180	11700	0.0081	-	-
1.6319	11800	0.0136	-	-
1.6457	11900	0.0147	-	-
1.6595	12000	0.0149	0.0121	0.4477
1.6734	12100	0.0143	-	-
1.6872	12200	0.0576	-	-
1.7010	12300	0.0355	-	-
1.7148	12400	0.0021	-	-
1.7287	12500	0.0158	-	-
1.7425	12600	0.0	-	-
1.7563	12700	0.0081	-	-
1.7702	12800	0.0012	-	-
1.7840	12900	0.0039	-	-
1.7978	13000	0.0203	0.0099	0.4580
1.8116	13100	0.0082	-	-
1.8255	13200	0.005	-	-
1.8393	13300	0.0109	-	-
1.8531	13400	0.0002	-	-
1.8670	13500	0.0067	-	-
1.8808	13600	0.0154	-	-
1.8946	13700	0.0021	-	-
1.9084	13800	0.0096	-	-
1.9223	13900	0.0064	-	-
1.9361	14000	0.006	0.0083	0.4691
1.9499	14100	0.0012	-	-
1.9638	14200	0.0018	-	-
1.9776	14300	0.0339	-	-
1.9914	14400	0.0191	-	-
2.0053	14500	0.0028	-	-
2.0191	14600	0.0068	-	-
2.0329	14700	0.0088	-	-
2.0467	14800	0.0625	-	-
2.0606	14900	0.0131	-	-
2.0744	15000	0.0052	0.0090	0.4483
2.0882	15100	0.0002	-	-
2.1021	15200	0.0108	-	-
2.1159	15300	0.0185	-	-
2.1297	15400	0.0002	-	-
2.1435	15500	0.0192	-	-
2.1574	15600	0.0082	-	-
2.1712	15700	0.0006	-	-
2.1850	15800	0.0095	-	-
2.1989	15900	0.0001	-	-
2.2127	16000	0.0136	0.0077	0.4718
2.2265	16100	0.009	-	-
2.2404	16200	0.0035	-	-
2.2542	16300	0.0001	-	-
2.2680	16400	0.008	-	-
2.2818	16500	0.0007	-	-
2.2957	16600	0.0123	-	-
2.3095	16700	0.0363	-	-
2.3233	16800	0.0034	-	-
2.3372	16900	0.0001	-	-
2.3510	17000	0.0219	0.0083	0.4428
2.3648	17100	0.0148	-	-
2.3786	17200	0.0	-	-
2.3925	17300	0.0005	-	-
2.4063	17400	0.0114	-	-
2.4201	17500	0.0367	-	-
2.4340	17600	0.0163	-	-
2.4478	17700	0.0083	-	-
2.4616	17800	0.0264	-	-
2.4755	17900	0.0059	-	-
2.4893	18000	0.001	0.0090	0.4408
2.5031	18100	0.0058	-	-
2.5169	18200	0.0002	-	-
2.5308	18300	0.0112	-	-
2.5446	18400	0.0194	-	-
2.5584	18500	0.0356	-	-
2.5723	18600	0.0136	-	-
2.5861	18700	0.0109	-	-
2.5999	18800	0.0184	-	-
2.6137	18900	0.0006	-	-
2.6276	19000	0.0094	0.0072	0.4510
2.6414	19100	0.0094	-	-
2.6552	19200	0.0007	-	-
2.6691	19300	0.0108	-	-
2.6829	19400	0.0123	-	-
2.6967	19500	0.0004	-	-
2.7106	19600	0.0004	-	-
2.7244	19700	0.0149	-	-
2.7382	19800	0.0	-	-
2.7520	19900	0.0	-	-
2.7659	20000	0.0005	0.0080	0.4617
2.7797	20100	0.0115	-	-
2.7935	20200	0.0	-	-
2.8074	20300	0.0	-	-
2.8212	20400	0.0017	-	-
2.8350	20500	0.0225	-	-
2.8488	20600	0.0251	-	-
2.8627	20700	0.0001	-	-
2.8765	20800	0.0013	-	-
2.8903	20900	0.0048	-	-
2.9042	21000	0.0016	0.0079	0.4548
2.9180	21100	0.0003	-	-
2.9318	21200	0.0352	-	-
2.9457	21300	0.0044	-	-
2.9595	21400	0.0124	-	-
2.9733	21500	0.0064	-	-
2.9871	21600	0.0086	-	-
3.0010	21700	0.0058	-	-
3.0148	21800	0.0018	-	-
3.0286	21900	0.0132	-	-
3.0425	22000	0.0144	0.0080	0.4472
3.0563	22100	0.0248	-	-
3.0701	22200	0.0139	-	-
3.0839	22300	0.0155	-	-
3.0978	22400	0.0115	-	-
3.1116	22500	0.0082	-	-
3.1254	22600	0.0068	-	-
3.1393	22700	0.0	-	-
3.1531	22800	0.0178	-	-
3.1669	22900	0.0007	-	-
3.1807	23000	0.0004	0.0072	0.4689
3.1946	23100	0.0	-	-
3.2084	23200	0.0	-	-
3.2222	23300	0.0128	-	-
3.2361	23400	0.0001	-	-
3.2499	23500	0.0027	-	-
3.2637	23600	0.0002	-	-
3.2776	23700	0.0048	-	-
3.2914	23800	0.0063	-	-
3.3052	23900	0.0331	-	-
3.3190	24000	0.0001	0.0089	0.4881
3.3329	24100	0.025	-	-
3.3467	24200	0.0045	-	-
3.3605	24300	0.0065	-	-
3.3744	24400	0.0003	-	-
3.3882	24500	0.0077	-	-
3.4020	24600	0.0002	-	-
3.4158	24700	0.0095	-	-
3.4297	24800	0.0219	-	-
3.4435	24900	0.0005	-	-
3.4573	25000	0.0114	0.0087	0.4686
3.4712	25100	0.0002	-	-
3.4850	25200	0.023	-	-
3.4988	25300	0.01	-	-
3.5127	25400	0.0114	-	-
3.5265	25500	0.0052	-	-
3.5403	25600	0.0095	-	-
3.5541	25700	0.0205	-	-
3.5680	25800	0.0002	-	-
3.5818	25900	0.0097	-	-
3.5956	26000	0.0207	0.0077	0.4741
3.6095	26100	0.0112	-	-
3.6233	26200	0.0045	-	-
3.6371	26300	0.0006	-	-
3.6509	26400	0.0302	-	-
3.6648	26500	0.007	-	-
3.6786	26600	0.0005	-	-
3.6924	26700	0.0086	-	-
3.7063	26800	0.0081	-	-
3.7201	26900	0.0006	-	-
3.7339	27000	0.0063	0.0099	0.4824
3.7478	27100	0.0198	-	-
3.7616	27200	0.0062	-	-
3.7754	27300	0.0	-	-
3.7892	27400	0.008	-	-
3.8031	27500	0.0034	-	-
3.8169	27600	0.0005	-	-
3.8307	27700	0.0065	-	-
3.8446	27800	0.0019	-	-
3.8584	27900	0.0108	-	-
3.8722	28000	0.0117	0.0069	0.4933
3.8860	28100	0.0106	-	-
3.8999	28200	0.0001	-	-
3.9137	28300	0.0	-	-
3.9275	28400	0.0066	-	-
3.9414	28500	0.011	-	-
3.9552	28600	0.0	-	-
3.9690	28700	0.0004	-	-
3.9829	28800	0.0081	-	-
3.9967	28900	0.0081	-	-
4.0105	29000	0.0122	0.0066	0.5047
4.0243	29100	0.0137	-	-
4.0382	29200	0.0098	-	-
4.0520	29300	0.0002	-	-
4.0658	29400	0.0075	-	-
4.0797	29500	0.0	-	-
4.0935	29600	0.0256	-	-
4.1073	29700	0.0096	-	-
4.1211	29800	0.0012	-	-
4.1350	29900	0.0048	-	-
4.1488	30000	0.0	0.0065	0.4963
4.1626	30100	0.0026	-	-
4.1765	30200	0.0025	-	-
4.1903	30300	0.0077	-	-
4.2041	30400	0.0168	-	-
4.2180	30500	0.0377	-	-
4.2318	30600	0.0	-	-
4.2456	30700	0.0114	-	-
4.2594	30800	0.0062	-	-
4.2733	30900	0.0135	-	-
4.2871	31000	0.0089	0.0080	0.4953
4.3009	31100	0.0106	-	-
4.3148	31200	0.0199	-	-
4.3286	31300	0.0066	-	-
4.3424	31400	0.0003	-	-
4.3562	31500	0.0045	-	-
4.3701	31600	0.0001	-	-
4.3839	31700	0.0064	-	-
4.3977	31800	0.0001	-	-
4.4116	31900	0.0052	-	-
4.4254	32000	0.011	0.0061	0.4994
4.4392	32100	0.0	-	-
4.4530	32200	0.015	-	-
4.4669	32300	0.0082	-	-
4.4807	32400	0.0	-	-
4.4945	32500	0.0041	-	-
4.5084	32600	0.0067	-	-
4.5222	32700	0.0003	-	-
4.5360	32800	0.0	-	-
4.5499	32900	0.002	-	-
4.5637	33000	0.0	0.0064	0.5035
4.5775	33100	0.0	-	-
4.5913	33200	0.0058	-	-
4.6052	33300	0.0033	-	-
4.6190	33400	0.008	-	-
4.6328	33500	0.0313	-	-
4.6467	33600	0.0294	-	-
4.6605	33700	0.0068	-	-
4.6743	33800	0.0068	-	-
4.6881	33900	0.0213	-	-
4.7020	34000	0.0117	0.0076	0.5044
4.7158	34100	0.0001	-	-
4.7296	34200	0.0024	-	-
4.7435	34300	0.0	-	-
4.7573	34400	0.0084	-	-
4.7711	34500	0.0091	-	-
4.7850	34600	0.0101	-	-
4.7988	34700	0.0093	-	-
4.8126	34800	0.0138	-	-
4.8264	34900	0.0113	-	-
4.8403	35000	0.0134	0.0064	0.5127
4.8541	35100	0.0233	-	-
4.8679	35200	0.0006	-	-
4.8818	35300	0.0	-	-
4.8956	35400	0.0095	-	-
4.9094	35500	0.0145	-	-
4.9232	35600	0.0075	-	-
4.9371	35700	0.0006	-	-
4.9509	35800	0.0	-	-
4.9647	35900	0.0	-	-
4.9786	36000	0.0136	0.0060	0.5170
4.9924	36100	0.0197	-	-
5.0062	36200	0.0127	-	-
5.0201	36300	0.0029	-	-
5.0339	36400	0.0028	-	-
5.0477	36500	0.011	-	-
5.0615	36600	0.0	-	-
5.0754	36700	0.0152	-	-
5.0892	36800	0.0076	-	-
5.1030	36900	0.0138	-	-
5.1169	37000	0.0002	0.0063	0.5164
5.1307	37100	0.0051	-	-
5.1445	37200	0.0158	-	-
5.1583	37300	0.0063	-	-
5.1722	37400	0.017	-	-
5.1860	37500	0.0115	-	-
5.1998	37600	0.0001	-	-
5.2137	37700	0.0072	-	-
5.2275	37800	0.0022	-	-
5.2413	37900	0.0045	-	-
5.2552	38000	0.0012	0.0052	0.5276
5.2690	38100	0.0	-	-
5.2828	38200	0.0127	-	-
5.2966	38300	0.006	-	-
5.3105	38400	0.0075	-	-
5.3243	38500	0.0	-	-
5.3381	38600	0.0001	-	-
5.3520	38700	0.0	-	-
5.3658	38800	0.0	-	-
5.3796	38900	0.0	-	-
5.3934	39000	0.0003	0.0053	0.5280
5.4073	39100	0.0	-	-
5.4211	39200	0.0	-	-
5.4349	39300	0.0	-	-
5.4488	39400	0.0002	-	-
5.4626	39500	0.0076	-	-
5.4764	39600	0.0016	-	-
5.4903	39700	0.0001	-	-
5.5041	39800	0.0	-	-
5.5179	39900	0.0	-	-
5.5317	40000	0.0	0.0052	0.5257
5.5456	40100	0.0081	-	-
5.5594	40200	0.0058	-	-
5.5732	40300	0.0067	-	-
5.5871	40400	0.007	-	-
5.6009	40500	0.0085	-	-
5.6147	40600	0.0015	-	-
5.6285	40700	0.0016	-	-
5.6424	40800	0.0007	-	-
5.6562	40900	0.0	-	-
5.6700	41000	0.0	0.0054	0.5337
5.6839	41100	0.0	-	-
5.6977	41200	0.0	-	-
5.7115	41300	0.0151	-	-
5.7253	41400	0.007	-	-
5.7392	41500	0.0	-	-
5.7530	41600	0.0052	-	-
5.7668	41700	0.0075	-	-
5.7807	41800	0.0099	-	-
5.7945	41900	0.0027	-	-
5.8083	42000	0.0001	0.0053	0.5346
5.8222	42100	0.0003	-	-
5.8360	42200	0.0	-	-
5.8498	42300	0.0	-	-
5.8636	42400	0.0055	-	-
5.8775	42500	0.0105	-	-
5.8913	42600	0.007	-	-
5.9051	42700	0.0001	-	-
5.9190	42800	0.0095	-	-
5.9328	42900	0.0075	-	-
5.9466	43000	0.0191	0.0052	0.5371
5.9604	43100	0.0002	-	-
5.9743	43200	0.0	-	-
5.9881	43300	0.004	-	-

Framework Versions

• Python: 3.11.13
• Sentence Transformers: 4.1.0
• Transformers: 4.52.4
• PyTorch: 2.6.0+cu124
• Accelerate: 1.8.1
• Datasets: 3.6.0
• Tokenizers: 0.21.2

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",
}

MultipleNegativesRankingLoss

@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}
}