ayushexel/emb-all-MiniLM-L6-v2-gooaq-6-epochs

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 on the gooaq dataset. 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.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: sentence-transformers/all-MiniLM-L6-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • gooaq
• Language: en

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: BertModel 
  (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})
  (2): Normalize()
)

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("ayushexel/emb-all-MiniLM-L6-v2-gooaq-6-epochs")
# Run inference
sentences = [
    'what is usb c ss?',
    'The USB Type-C specification is pretty confusing. ... The standard USB logo to identify USB 2.0 ports or slower. "SS" markings, which stand for SuperSpeed, to identify USB 3.0 ports, otherwise known as USB 3.1 gen 1. "10" markings, which stand for 10 Gbps, to identify USB 3.1 gen 2 ports with ultra-fast connectivity.',
    '“Global warming” refers to the rise in global temperatures due mainly to the increasing concentrations of greenhouse gases in the atmosphere. “Climate change” refers to the increasing changes in the measures of climate over a long period of time – including precipitation, temperature, and wind patterns.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Triplet

• Dataset: gooqa-dev
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.6018

Training Details

Training Dataset

gooaq

• Dataset: gooaq at b089f72
• Size: 1,995,000 training samples
• Columns: question and answer
• Approximate statistics based on the first 1000 samples:

  	question	answer
  type	string	string
  details	min: 8 tokens mean: 11.86 tokens max: 23 tokens	min: 14 tokens mean: 60.74 tokens max: 133 tokens

• Samples:

  question	answer
  can twine be a noun?	noun. a strong thread or string composed of two or more strands twisted together. an act of twining, twisting, or interweaving.
  what is bo id in nsdl?	The demat account number allotted to the beneficiary holder(s) by DP is known as the BO-ID. In CDSL it is 16 digits number. It is an intermediary (an institution) between the investor and the depository.
  how much does it cost to run an electric fan all night?	The average indoor ceiling fan costs around 0.13c to 1.29c per hour to run, or between $1.90 and $18.85 each year. This will depend on the fan's speed settings, how frequently it's used, and the rate you pay on electricity. Like most electrical appliances, a ceiling fan's power is measured in watts.

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

gooaq

• Dataset: gooaq at b089f72
• Size: 5,000 evaluation samples
• Columns: question and answer
• Approximate statistics based on the first 1000 samples:

  	question	answer
  type	string	string
  details	min: 8 tokens mean: 11.8 tokens max: 21 tokens	min: 14 tokens mean: 60.68 tokens max: 123 tokens

• Samples:

  question	answer
  how much water should a person drink in 8 hours?	Health authorities commonly recommend eight 8-ounce glasses, which equals about 2 liters, or half a gallon. This is called the 8×8 rule and is very easy to remember. However, some health gurus believe that you need to sip on water constantly throughout the day, even when you're not thirsty.
  what does this mean in excel #name?	Important: The #NAME? error signifies that something needs to be corrected in the syntax, so when you see the error in your formula, resolve it. Do not use any error-handling functions such as IFERROR to mask the error. To avoid typos in formula names, use the Formula Wizard in Excel.
  are hydroflask good for the environment?	Hydro Flasks are a new fad among many students and adults to help minimize plastic waste in the oceans. Hydro Flasks are great because they use a type of metal called TempShield, which keeps your beverage or food either hot for up to six hours or cold for up to twenty-four hours.

• 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: 256
• per_device_eval_batch_size: 256
• 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: 256
• per_device_eval_batch_size: 256
• 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: 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: 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}
• 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
• 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_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	gooqa-dev_cosine_accuracy
-1	-1	-	-	0.5378
0.0128	100	0.0766	-	-
0.0257	200	0.0733	-	-
0.0385	300	0.0697	-	-
0.0513	400	0.0689	-	-
0.0642	500	0.0693	-	-
0.0770	600	0.07	-	-
0.0898	700	0.0672	-	-
0.1027	800	0.0636	-	-
0.1155	900	0.0664	-	-
0.1283	1000	0.0656	0.0454	0.5484
0.1412	1100	0.0648	-	-
0.1540	1200	0.0643	-	-
0.1668	1300	0.0613	-	-
0.1796	1400	0.0617	-	-
0.1925	1500	0.0609	-	-
0.2053	1600	0.0622	-	-
0.2181	1700	0.0628	-	-
0.2310	1800	0.0616	-	-
0.2438	1900	0.0632	-	-
0.2566	2000	0.0614	0.0405	0.5446
0.2695	2100	0.0617	-	-
0.2823	2200	0.0577	-	-
0.2951	2300	0.0579	-	-
0.3080	2400	0.0599	-	-
0.3208	2500	0.056	-	-
0.3336	2600	0.0574	-	-
0.3465	2700	0.0595	-	-
0.3593	2800	0.0581	-	-
0.3721	2900	0.0588	-	-
0.3850	3000	0.0576	0.0412	0.5386
0.3978	3100	0.0613	-	-
0.4106	3200	0.0556	-	-
0.4235	3300	0.0575	-	-
0.4363	3400	0.0595	-	-
0.4491	3500	0.0559	-	-
0.4620	3600	0.0558	-	-
0.4748	3700	0.0544	-	-
0.4876	3800	0.0554	-	-
0.5004	3900	0.0581	-	-
0.5133	4000	0.0546	0.0406	0.5462
0.5261	4100	0.0559	-	-
0.5389	4200	0.0535	-	-
0.5518	4300	0.0533	-	-
0.5646	4400	0.0565	-	-
0.5774	4500	0.0562	-	-
0.5903	4600	0.0555	-	-
0.6031	4700	0.055	-	-
0.6159	4800	0.0548	-	-
0.6288	4900	0.0567	-	-
0.6416	5000	0.0564	0.0374	0.5444
0.6544	5100	0.0541	-	-
0.6673	5200	0.0538	-	-
0.6801	5300	0.0541	-	-
0.6929	5400	0.0585	-	-
0.7058	5500	0.0539	-	-
0.7186	5600	0.0551	-	-
0.7314	5700	0.0534	-	-
0.7443	5800	0.0517	-	-
0.7571	5900	0.0525	-	-
0.7699	6000	0.0503	0.0380	0.5482
0.7828	6100	0.0541	-	-
0.7956	6200	0.0526	-	-
0.8084	6300	0.0498	-	-
0.8212	6400	0.0544	-	-
0.8341	6500	0.0511	-	-
0.8469	6600	0.0539	-	-
0.8597	6700	0.05	-	-
0.8726	6800	0.05	-	-
0.8854	6900	0.0508	-	-
0.8982	7000	0.0498	0.0347	0.5558
0.9111	7100	0.0481	-	-
0.9239	7200	0.0509	-	-
0.9367	7300	0.0481	-	-
0.9496	7400	0.0487	-	-
0.9624	7500	0.0508	-	-
0.9752	7600	0.0477	-	-
0.9881	7700	0.0482	-	-
1.0009	7800	0.0469	-	-
1.0137	7900	0.0436	-	-
1.0266	8000	0.042	0.0333	0.5560
1.0394	8100	0.0408	-	-
1.0522	8200	0.0417	-	-
1.0651	8300	0.0449	-	-
1.0779	8400	0.042	-	-
1.0907	8500	0.0425	-	-
1.1036	8600	0.043	-	-
1.1164	8700	0.0389	-	-
1.1292	8800	0.043	-	-
1.1421	8900	0.0419	-	-
1.1549	9000	0.0418	0.0336	0.5556
1.1677	9100	0.0435	-	-
1.1805	9200	0.0404	-	-
1.1934	9300	0.0406	-	-
1.2062	9400	0.0406	-	-
1.2190	9500	0.0431	-	-
1.2319	9600	0.0385	-	-
1.2447	9700	0.0428	-	-
1.2575	9800	0.0424	-	-
1.2704	9900	0.0412	-	-
1.2832	10000	0.0389	0.0336	0.5688
1.2960	10100	0.0382	-	-
1.3089	10200	0.0423	-	-
1.3217	10300	0.0423	-	-
1.3345	10400	0.039	-	-
1.3474	10500	0.0404	-	-
1.3602	10600	0.0383	-	-
1.3730	10700	0.0416	-	-
1.3859	10800	0.0403	-	-
1.3987	10900	0.0408	-	-
1.4115	11000	0.0388	0.0301	0.5752
1.4244	11100	0.039	-	-
1.4372	11200	0.0391	-	-
1.4500	11300	0.0399	-	-
1.4629	11400	0.0387	-	-
1.4757	11500	0.0387	-	-
1.4885	11600	0.0395	-	-
1.5013	11700	0.0424	-	-
1.5142	11800	0.0382	-	-
1.5270	11900	0.0393	-	-
1.5398	12000	0.04	0.0320	0.5664
1.5527	12100	0.0369	-	-
1.5655	12200	0.0385	-	-
1.5783	12300	0.0414	-	-
1.5912	12400	0.0406	-	-
1.6040	12500	0.0407	-	-
1.6168	12600	0.0378	-	-
1.6297	12700	0.0393	-	-
1.6425	12800	0.0389	-	-
1.6553	12900	0.0407	-	-
1.6682	13000	0.0413	0.0300	0.5674
1.6810	13100	0.0383	-	-
1.6938	13200	0.0374	-	-
1.7067	13300	0.0388	-	-
1.7195	13400	0.0395	-	-
1.7323	13500	0.0381	-	-
1.7452	13600	0.039	-	-
1.7580	13700	0.0403	-	-
1.7708	13800	0.0381	-	-
1.7837	13900	0.0389	-	-
1.7965	14000	0.0387	0.0289	0.5726
1.8093	14100	0.0402	-	-
1.8221	14200	0.0375	-	-
1.8350	14300	0.0355	-	-
1.8478	14400	0.0383	-	-
1.8606	14500	0.037	-	-
1.8735	14600	0.0363	-	-
1.8863	14700	0.0361	-	-
1.8991	14800	0.0375	-	-
1.9120	14900	0.0381	-	-
1.9248	15000	0.0387	0.0292	0.5776
1.9376	15100	0.0381	-	-
1.9505	15200	0.0385	-	-
1.9633	15300	0.0359	-	-
1.9761	15400	0.039	-	-
1.9890	15500	0.0379	-	-
2.0018	15600	0.0364	-	-
2.0146	15700	0.0302	-	-
2.0275	15800	0.033	-	-
2.0403	15900	0.0317	-	-
2.0531	16000	0.0303	0.0276	0.5712
2.0660	16100	0.0311	-	-
2.0788	16200	0.0328	-	-
2.0916	16300	0.032	-	-
2.1045	16400	0.0304	-	-
2.1173	16500	0.0292	-	-
2.1301	16600	0.0302	-	-
2.1429	16700	0.0326	-	-
2.1558	16800	0.0303	-	-
2.1686	16900	0.0309	-	-
2.1814	17000	0.0315	0.0279	0.5760
2.1943	17100	0.0329	-	-
2.2071	17200	0.0303	-	-
2.2199	17300	0.0318	-	-
2.2328	17400	0.0312	-	-
2.2456	17500	0.0321	-	-
2.2584	17600	0.0303	-	-
2.2713	17700	0.0314	-	-
2.2841	17800	0.0297	-	-
2.2969	17900	0.03	-	-
2.3098	18000	0.0317	0.0268	0.5898
2.3226	18100	0.0333	-	-
2.3354	18200	0.0286	-	-
2.3483	18300	0.032	-	-
2.3611	18400	0.0311	-	-
2.3739	18500	0.0298	-	-
2.3868	18600	0.0296	-	-
2.3996	18700	0.032	-	-
2.4124	18800	0.0301	-	-
2.4253	18900	0.0303	-	-
2.4381	19000	0.0296	0.0252	0.5816
2.4509	19100	0.0304	-	-
2.4637	19200	0.0311	-	-
2.4766	19300	0.0306	-	-
2.4894	19400	0.0309	-	-
2.5022	19500	0.0312	-	-
2.5151	19600	0.0284	-	-
2.5279	19700	0.0296	-	-
2.5407	19800	0.0309	-	-
2.5536	19900	0.0305	-	-
2.5664	20000	0.0289	0.0264	0.5888
2.5792	20100	0.03	-	-
2.5921	20200	0.0274	-	-
2.6049	20300	0.0296	-	-
2.6177	20400	0.0297	-	-
2.6306	20500	0.0303	-	-
2.6434	20600	0.0324	-	-
2.6562	20700	0.0309	-	-
2.6691	20800	0.031	-	-
2.6819	20900	0.0286	-	-
2.6947	21000	0.0286	0.0258	0.5854
2.7076	21100	0.0289	-	-
2.7204	21200	0.0294	-	-
2.7332	21300	0.0301	-	-
2.7461	21400	0.0289	-	-
2.7589	21500	0.0318	-	-
2.7717	21600	0.0288	-	-
2.7846	21700	0.0291	-	-
2.7974	21800	0.0298	-	-
2.8102	21900	0.0297	-	-
2.8230	22000	0.0284	0.0266	0.5864
2.8359	22100	0.0294	-	-
2.8487	22200	0.0284	-	-
2.8615	22300	0.031	-	-
2.8744	22400	0.0294	-	-
2.8872	22500	0.0301	-	-
2.9000	22600	0.0293	-	-
2.9129	22700	0.0296	-	-
2.9257	22800	0.029	-	-
2.9385	22900	0.0292	-	-
2.9514	23000	0.0305	0.0256	0.5876
2.9642	23100	0.0296	-	-
2.9770	23200	0.0299	-	-
2.9899	23300	0.0316	-	-
3.0027	23400	0.0273	-	-
3.0155	23500	0.0252	-	-
3.0284	23600	0.0249	-	-
3.0412	23700	0.0237	-	-
3.0540	23800	0.0242	-	-
3.0669	23900	0.0256	-	-
3.0797	24000	0.0245	0.0245	0.5834
3.0925	24100	0.0248	-	-
3.1054	24200	0.0251	-	-
3.1182	24300	0.026	-	-
3.1310	24400	0.0261	-	-
3.1438	24500	0.0243	-	-
3.1567	24600	0.0243	-	-
3.1695	24700	0.0256	-	-
3.1823	24800	0.0262	-	-
3.1952	24900	0.0261	-	-
3.2080	25000	0.0265	0.0249	0.5816
3.2208	25100	0.0276	-	-
3.2337	25200	0.0237	-	-
3.2465	25300	0.0285	-	-
3.2593	25400	0.0234	-	-
3.2722	25500	0.0252	-	-
3.2850	25600	0.0249	-	-
3.2978	25700	0.0245	-	-
3.3107	25800	0.0254	-	-
3.3235	25900	0.025	-	-
3.3363	26000	0.0273	0.0243	0.5914
3.3492	26100	0.0234	-	-
3.3620	26200	0.0247	-	-
3.3748	26300	0.0244	-	-
3.3877	26400	0.0242	-	-
3.4005	26500	0.0264	-	-
3.4133	26600	0.026	-	-
3.4262	26700	0.0259	-	-
3.4390	26800	0.0253	-	-
3.4518	26900	0.0251	-	-
3.4646	27000	0.0249	0.0234	0.5950
3.4775	27100	0.0257	-	-
3.4903	27200	0.0259	-	-
3.5031	27300	0.0262	-	-
3.5160	27400	0.0252	-	-
3.5288	27500	0.0268	-	-
3.5416	27600	0.0245	-	-
3.5545	27700	0.026	-	-
3.5673	27800	0.0237	-	-
3.5801	27900	0.0261	-	-
3.5930	28000	0.0256	0.0226	0.5944
3.6058	28100	0.0256	-	-
3.6186	28200	0.0243	-	-
3.6315	28300	0.0262	-	-
3.6443	28400	0.0265	-	-
3.6571	28500	0.0237	-	-
3.6700	28600	0.0243	-	-
3.6828	28700	0.0247	-	-
3.6956	28800	0.024	-	-
3.7085	28900	0.0259	-	-
3.7213	29000	0.0268	0.0238	0.5942
3.7341	29100	0.0255	-	-
3.7470	29200	0.0249	-	-
3.7598	29300	0.0244	-	-
3.7726	29400	0.0258	-	-
3.7854	29500	0.0253	-	-
3.7983	29600	0.0246	-	-
3.8111	29700	0.0245	-	-
3.8239	29800	0.0269	-	-
3.8368	29900	0.0256	-	-
3.8496	30000	0.0242	0.0236	0.5976
3.8624	30100	0.0235	-	-
3.8753	30200	0.025	-	-
3.8881	30300	0.024	-	-
3.9009	30400	0.0247	-	-
3.9138	30500	0.0261	-	-
3.9266	30600	0.0251	-	-
3.9394	30700	0.0233	-	-
3.9523	30800	0.0239	-	-
3.9651	30900	0.0252	-	-
3.9779	31000	0.0252	0.0228	0.5956
3.9908	31100	0.0236	-	-
4.0036	31200	0.0253	-	-
4.0164	31300	0.0223	-	-
4.0293	31400	0.0224	-	-
4.0421	31500	0.0229	-	-
4.0549	31600	0.0234	-	-
4.0678	31700	0.0223	-	-
4.0806	31800	0.0225	-	-
4.0934	31900	0.0239	-	-
4.1062	32000	0.0221	0.0241	0.5892
4.1191	32100	0.0227	-	-
4.1319	32200	0.0232	-	-
4.1447	32300	0.0214	-	-
4.1576	32400	0.0224	-	-
4.1704	32500	0.0229	-	-
4.1832	32600	0.0214	-	-
4.1961	32700	0.0237	-	-
4.2089	32800	0.0213	-	-
4.2217	32900	0.0212	-	-
4.2346	33000	0.0213	0.0230	0.5938
4.2474	33100	0.0225	-	-
4.2602	33200	0.0237	-	-
4.2731	33300	0.0236	-	-
4.2859	33400	0.0217	-	-
4.2987	33500	0.0212	-	-
4.3116	33600	0.0227	-	-
4.3244	33700	0.0223	-	-
4.3372	33800	0.0211	-	-
4.3501	33900	0.0224	-	-
4.3629	34000	0.0231	0.0223	0.6000
4.3757	34100	0.0218	-	-
4.3886	34200	0.0227	-	-
4.4014	34300	0.0223	-	-
4.4142	34400	0.0227	-	-
4.4270	34500	0.022	-	-
4.4399	34600	0.021	-	-
4.4527	34700	0.0203	-	-
4.4655	34800	0.0219	-	-
4.4784	34900	0.0221	-	-
4.4912	35000	0.0213	0.0219	0.5924
4.5040	35100	0.0229	-	-
4.5169	35200	0.0217	-	-
4.5297	35300	0.0207	-	-
4.5425	35400	0.022	-	-
4.5554	35500	0.0228	-	-
4.5682	35600	0.0214	-	-
4.5810	35700	0.0225	-	-
4.5939	35800	0.0209	-	-
4.6067	35900	0.0214	-	-
4.6195	36000	0.0208	0.0218	0.5994
4.6324	36100	0.021	-	-
4.6452	36200	0.0198	-	-
4.6580	36300	0.021	-	-
4.6709	36400	0.022	-	-
4.6837	36500	0.0209	-	-
4.6965	36600	0.0207	-	-
4.7094	36700	0.0212	-	-
4.7222	36800	0.0219	-	-
4.7350	36900	0.0218	-	-
4.7479	37000	0.0233	0.0217	0.5984
4.7607	37100	0.0227	-	-
4.7735	37200	0.0218	-	-
4.7863	37300	0.023	-	-
4.7992	37400	0.0228	-	-
4.8120	37500	0.0217	-	-
4.8248	37600	0.0214	-	-
4.8377	37700	0.0225	-	-
4.8505	37800	0.0214	-	-
4.8633	37900	0.0194	-	-
4.8762	38000	0.0217	0.0217	0.6020
4.8890	38100	0.0218	-	-
4.9018	38200	0.0225	-	-
4.9147	38300	0.0218	-	-
4.9275	38400	0.021	-	-
4.9403	38500	0.0221	-	-
4.9532	38600	0.0239	-	-
4.9660	38700	0.0213	-	-
4.9788	38800	0.0218	-	-
4.9917	38900	0.0219	-	-
5.0045	39000	0.0214	0.0206	0.5936
5.0173	39100	0.02	-	-
5.0302	39200	0.0202	-	-
5.0430	39300	0.02	-	-
5.0558	39400	0.0198	-	-
5.0687	39500	0.0189	-	-
5.0815	39600	0.0207	-	-
5.0943	39700	0.0201	-	-
5.1071	39800	0.0211	-	-
5.1200	39900	0.0209	-	-
5.1328	40000	0.0206	0.0213	0.6004
5.1456	40100	0.0194	-	-
5.1585	40200	0.0205	-	-
5.1713	40300	0.0195	-	-
5.1841	40400	0.0208	-	-
5.1970	40500	0.0196	-	-
5.2098	40600	0.0211	-	-
5.2226	40700	0.019	-	-
5.2355	40800	0.0208	-	-
5.2483	40900	0.02	-	-
5.2611	41000	0.0213	0.0209	0.6006
5.2740	41100	0.0195	-	-
5.2868	41200	0.0208	-	-
5.2996	41300	0.0215	-	-
5.3125	41400	0.0195	-	-
5.3253	41500	0.0194	-	-
5.3381	41600	0.0201	-	-
5.3510	41700	0.0208	-	-
5.3638	41800	0.0192	-	-
5.3766	41900	0.0205	-	-
5.3895	42000	0.0201	0.0211	0.6012
5.4023	42100	0.0199	-	-
5.4151	42200	0.0183	-	-
5.4279	42300	0.0184	-	-
5.4408	42400	0.0196	-	-
5.4536	42500	0.0191	-	-
5.4664	42600	0.0193	-	-
5.4793	42700	0.019	-	-
5.4921	42800	0.0205	-	-
5.5049	42900	0.0193	-	-
5.5178	43000	0.0189	0.0210	0.6028
5.5306	43100	0.0213	-	-
5.5434	43200	0.0203	-	-
5.5563	43300	0.0206	-	-
5.5691	43400	0.0193	-	-
5.5819	43500	0.019	-	-
5.5948	43600	0.0186	-	-
5.6076	43700	0.0192	-	-
5.6204	43800	0.0198	-	-
5.6333	43900	0.0207	-	-
5.6461	44000	0.0208	0.0208	0.6038
5.6589	44100	0.0188	-	-
5.6718	44200	0.0182	-	-
5.6846	44300	0.0196	-	-
5.6974	44400	0.0198	-	-
5.7103	44500	0.0206	-	-
5.7231	44600	0.0205	-	-
5.7359	44700	0.0195	-	-
5.7487	44800	0.0191	-	-
5.7616	44900	0.0198	-	-
5.7744	45000	0.0202	0.0208	0.6054
5.7872	45100	0.0198	-	-
5.8001	45200	0.018	-	-
5.8129	45300	0.0199	-	-
5.8257	45400	0.0191	-	-
5.8386	45500	0.0202	-	-
5.8514	45600	0.0195	-	-
5.8642	45700	0.0194	-	-
5.8771	45800	0.0195	-	-
5.8899	45900	0.0189	-	-
5.9027	46000	0.0199	0.0208	0.6044
5.9156	46100	0.0192	-	-
5.9284	46200	0.0204	-	-
5.9412	46300	0.0187	-	-
5.9541	46400	0.0184	-	-
5.9669	46500	0.0202	-	-
5.9797	46600	0.0191	-	-
5.9926	46700	0.0196	-	-
-1	-1	-	-	0.6018

Framework Versions

• Python: 3.11.0
• Sentence Transformers: 4.0.1
• Transformers: 4.50.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.5.2
• Datasets: 3.5.0
• Tokenizers: 0.21.1

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}
}