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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("ayushexel/emb-all-MiniLM-L6-v2-gooaq-10-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	0.6154

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:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim" 4}

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:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim" 4}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 256
per_device_eval_batch_size: 256
num_train_epochs: 10
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: 10
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	-	-	0.5350
0.0128	100	0.0767	-	-
0.0257	200	0.0738	-	-
0.0385	300	0.0707	-	-
0.0513	400	0.0701	-	-
0.0642	500	0.0707	-	-
0.0770	600	0.0714	-	-
0.0898	700	0.0683	-	-
0.1027	800	0.0649	-	-
0.1155	900	0.0677	-	-
0.1283	1000	0.067	0.0465	0.5472
0.1412	1100	0.066	-	-
0.1540	1200	0.0655	-	-
0.1668	1300	0.0624	-	-
0.1796	1400	0.0629	-	-
0.1925	1500	0.062	-	-
0.2053	1600	0.0633	-	-
0.2181	1700	0.0638	-	-
0.2310	1800	0.0625	-	-
0.2438	1900	0.0643	-	-
0.2566	2000	0.0623	0.0416	0.5496
0.2695	2100	0.0627	-	-
0.2823	2200	0.0586	-	-
0.2951	2300	0.0585	-	-
0.3080	2400	0.0606	-	-
0.3208	2500	0.0567	-	-
0.3336	2600	0.0581	-	-
0.3465	2700	0.0602	-	-
0.3593	2800	0.0584	-	-
0.3721	2900	0.0591	-	-
0.3850	3000	0.0582	0.0414	0.5392
0.3978	3100	0.061	-	-
0.4106	3200	0.056	-	-
0.4235	3300	0.0577	-	-
0.4363	3400	0.0593	-	-
0.4491	3500	0.056	-	-
0.4620	3600	0.0557	-	-
0.4748	3700	0.0542	-	-
0.4876	3800	0.0552	-	-
0.5004	3900	0.0577	-	-
0.5133	4000	0.0545	0.0398	0.5518
0.5261	4100	0.0558	-	-
0.5389	4200	0.0531	-	-
0.5518	4300	0.0529	-	-
0.5646	4400	0.0558	-	-
0.5774	4500	0.0555	-	-
0.5903	4600	0.0547	-	-
0.6031	4700	0.0542	-	-
0.6159	4800	0.0539	-	-
0.6288	4900	0.0559	-	-
0.6416	5000	0.0555	0.0370	0.5502
0.6544	5100	0.0539	-	-
0.6673	5200	0.0536	-	-
0.6801	5300	0.0533	-	-
0.6929	5400	0.0584	-	-
0.7058	5500	0.0537	-	-
0.7186	5600	0.0553	-	-
0.7314	5700	0.0537	-	-
0.7443	5800	0.052	-	-
0.7571	5900	0.0532	-	-
0.7699	6000	0.0509	0.0379	0.5484
0.7828	6100	0.0545	-	-
0.7956	6200	0.0534	-	-
0.8084	6300	0.0502	-	-
0.8212	6400	0.0555	-	-
0.8341	6500	0.0524	-	-
0.8469	6600	0.0553	-	-
0.8597	6700	0.0514	-	-
0.8726	6800	0.0516	-	-
0.8854	6900	0.0522	-	-
0.8982	7000	0.0514	0.0360	0.5582
0.9111	7100	0.0501	-	-
0.9239	7200	0.0529	-	-
0.9367	7300	0.05	-	-
0.9496	7400	0.0509	-	-
0.9624	7500	0.053	-	-
0.9752	7600	0.0503	-	-
0.9881	7700	0.0508	-	-
1.0009	7800	0.0492	-	-
1.0137	7900	0.0475	-	-
1.0266	8000	0.0455	0.0358	0.5482
1.0394	8100	0.0447	-	-
1.0522	8200	0.0454	-	-
1.0651	8300	0.0488	-	-
1.0779	8400	0.0458	-	-
1.0907	8500	0.0461	-	-
1.1036	8600	0.0468	-	-
1.1164	8700	0.0427	-	-
1.1292	8800	0.047	-	-
1.1421	8900	0.0456	-	-
1.1549	9000	0.0455	0.0357	0.5506
1.1677	9100	0.0472	-	-
1.1805	9200	0.0441	-	-
1.1934	9300	0.0445	-	-
1.2062	9400	0.0445	-	-
1.2190	9500	0.0473	-	-
1.2319	9600	0.042	-	-
1.2447	9700	0.0465	-	-
1.2575	9800	0.0456	-	-
1.2704	9900	0.0446	-	-
1.2832	10000	0.0424	0.0355	0.5664
1.2960	10100	0.0417	-	-
1.3089	10200	0.0457	-	-
1.3217	10300	0.0458	-	-
1.3345	10400	0.0423	-	-
1.3474	10500	0.0438	-	-
1.3602	10600	0.0413	-	-
1.3730	10700	0.0452	-	-
1.3859	10800	0.0436	-	-
1.3987	10900	0.044	-	-
1.4115	11000	0.0422	0.0319	0.5680
1.4244	11100	0.0424	-	-
1.4372	11200	0.0425	-	-
1.4500	11300	0.0428	-	-
1.4629	11400	0.0416	-	-
1.4757	11500	0.042	-	-
1.4885	11600	0.0429	-	-
1.5013	11700	0.0459	-	-
1.5142	11800	0.0411	-	-
1.5270	11900	0.0424	-	-
1.5398	12000	0.0431	0.0336	0.5628
1.5527	12100	0.0398	-	-
1.5655	12200	0.0416	-	-
1.5783	12300	0.0444	-	-
1.5912	12400	0.0438	-	-
1.6040	12500	0.0439	-	-
1.6168	12600	0.0408	-	-
1.6297	12700	0.0426	-	-
1.6425	12800	0.0417	-	-
1.6553	12900	0.044	-	-
1.6682	13000	0.0444	0.0314	0.5670
1.6810	13100	0.0414	-	-
1.6938	13200	0.0403	-	-
1.7067	13300	0.0417	-	-
1.7195	13400	0.0426	-	-
1.7323	13500	0.041	-	-
1.7452	13600	0.0422	-	-
1.7580	13700	0.0435	-	-
1.7708	13800	0.0413	-	-
1.7837	13900	0.0419	-	-
1.7965	14000	0.0415	0.0302	0.5692
1.8093	14100	0.0432	-	-
1.8221	14200	0.0404	-	-
1.8350	14300	0.0382	-	-
1.8478	14400	0.0412	-	-
1.8606	14500	0.0396	-	-
1.8735	14600	0.039	-	-
1.8863	14700	0.039	-	-
1.8991	14800	0.0403	-	-
1.9120	14900	0.0406	-	-
1.9248	15000	0.0416	0.0309	0.5752
1.9376	15100	0.0411	-	-
1.9505	15200	0.0418	-	-
1.9633	15300	0.0386	-	-
1.9761	15400	0.0423	-	-
1.9890	15500	0.0408	-	-
2.0018	15600	0.0389	-	-
2.0146	15700	0.032	-	-
2.0275	15800	0.035	-	-
2.0403	15900	0.0337	-	-
2.0531	16000	0.0321	0.0288	0.5704
2.0660	16100	0.0329	-	-
2.0788	16200	0.0346	-	-
2.0916	16300	0.034	-	-
2.1045	16400	0.0319	-	-
2.1173	16500	0.0313	-	-
2.1301	16600	0.032	-	-
2.1429	16700	0.0346	-	-
2.1558	16800	0.0322	-	-
2.1686	16900	0.033	-	-
2.1814	17000	0.0336	0.0290	0.5768
2.1943	17100	0.0351	-	-
2.2071	17200	0.0324	-	-
2.2199	17300	0.0337	-	-
2.2328	17400	0.0332	-	-
2.2456	17500	0.0339	-	-
2.2584	17600	0.0324	-	-
2.2713	17700	0.0335	-	-
2.2841	17800	0.0316	-	-
2.2969	17900	0.0317	-	-
2.3098	18000	0.0338	0.0271	0.5872
2.3226	18100	0.0354	-	-
2.3354	18200	0.0305	-	-
2.3483	18300	0.0339	-	-
2.3611	18400	0.0333	-	-
2.3739	18500	0.0317	-	-
2.3868	18600	0.0316	-	-
2.3996	18700	0.0342	-	-
2.4124	18800	0.032	-	-
2.4253	18900	0.0325	-	-
2.4381	19000	0.0317	0.0262	0.5768
2.4509	19100	0.0324	-	-
2.4637	19200	0.0331	-	-
2.4766	19300	0.0326	-	-
2.4894	19400	0.0332	-	-
2.5022	19500	0.0332	-	-
2.5151	19600	0.0302	-	-
2.5279	19700	0.0317	-	-
2.5407	19800	0.0331	-	-
2.5536	19900	0.0325	-	-
2.5664	20000	0.0309	0.0275	0.5854
2.5792	20100	0.0322	-	-
2.5921	20200	0.0292	-	-
2.6049	20300	0.0319	-	-
2.6177	20400	0.0317	-	-
2.6306	20500	0.0324	-	-
2.6434	20600	0.0343	-	-
2.6562	20700	0.033	-	-
2.6691	20800	0.0332	-	-
2.6819	20900	0.0307	-	-
2.6947	21000	0.0307	0.0270	0.5800
2.7076	21100	0.0312	-	-
2.7204	21200	0.0317	-	-
2.7332	21300	0.0322	-	-
2.7461	21400	0.031	-	-
2.7589	21500	0.034	-	-
2.7717	21600	0.0308	-	-
2.7846	21700	0.0314	-	-
2.7974	21800	0.0319	-	-
2.8102	21900	0.0317	-	-
2.8230	22000	0.0308	0.0285	0.5820
2.8359	22100	0.0317	-	-
2.8487	22200	0.0304	-	-
2.8615	22300	0.0332	-	-
2.8744	22400	0.0317	-	-
2.8872	22500	0.0324	-	-
2.9000	22600	0.0316	-	-
2.9129	22700	0.0316	-	-
2.9257	22800	0.0314	-	-
2.9385	22900	0.0313	-	-
2.9514	23000	0.0327	0.0267	0.5822
2.9642	23100	0.0317	-	-
2.9770	23200	0.0321	-	-
2.9899	23300	0.0341	-	-
3.0027	23400	0.0293	-	-
3.0155	23500	0.0263	-	-
3.0284	23600	0.0263	-	-
3.0412	23700	0.0248	-	-
3.0540	23800	0.0254	-	-
3.0669	23900	0.0269	-	-
3.0797	24000	0.0257	0.0258	0.5816
3.0925	24100	0.026	-	-
3.1054	24200	0.026	-	-
3.1182	24300	0.0275	-	-
3.1310	24400	0.0274	-	-
3.1438	24500	0.0257	-	-
3.1567	24600	0.0256	-	-
3.1695	24700	0.027	-	-
3.1823	24800	0.0277	-	-
3.1952	24900	0.0278	-	-
3.2080	25000	0.0281	0.0263	0.5788
3.2208	25100	0.0292	-	-
3.2337	25200	0.025	-	-
3.2465	25300	0.03	-	-
3.2593	25400	0.0251	-	-
3.2722	25500	0.0266	-	-
3.2850	25600	0.0262	-	-
3.2978	25700	0.0259	-	-
3.3107	25800	0.0268	-	-
3.3235	25900	0.0266	-	-
3.3363	26000	0.0291	0.0253	0.5902
3.3492	26100	0.0248	-	-
3.3620	26200	0.0263	-	-
3.3748	26300	0.0262	-	-
3.3877	26400	0.0258	-	-
3.4005	26500	0.0282	-	-
3.4133	26600	0.0278	-	-
3.4262	26700	0.0273	-	-
3.4390	26800	0.027	-	-
3.4518	26900	0.0266	-	-
3.4646	27000	0.0267	0.0250	0.5932
3.4775	27100	0.0274	-	-
3.4903	27200	0.0276	-	-
3.5031	27300	0.0282	-	-
3.5160	27400	0.0271	-	-
3.5288	27500	0.0282	-	-
3.5416	27600	0.0259	-	-
3.5545	27700	0.0278	-	-
3.5673	27800	0.0254	-	-
3.5801	27900	0.0279	-	-
3.5930	28000	0.0272	0.0243	0.5932
3.6058	28100	0.0272	-	-
3.6186	28200	0.0259	-	-
3.6315	28300	0.0279	-	-
3.6443	28400	0.0283	-	-
3.6571	28500	0.0256	-	-
3.6700	28600	0.0262	-	-
3.6828	28700	0.0265	-	-
3.6956	28800	0.0254	-	-
3.7085	28900	0.0276	-	-
3.7213	29000	0.0292	0.0255	0.5956
3.7341	29100	0.0273	-	-
3.7470	29200	0.0272	-	-
3.7598	29300	0.0263	-	-
3.7726	29400	0.0275	-	-
3.7854	29500	0.0272	-	-
3.7983	29600	0.0262	-	-
3.8111	29700	0.0266	-	-
3.8239	29800	0.0288	-	-
3.8368	29900	0.0276	-	-
3.8496	30000	0.0261	0.0249	0.6028
3.8624	30100	0.0254	-	-
3.8753	30200	0.0267	-	-
3.8881	30300	0.0262	-	-
3.9009	30400	0.026	-	-
3.9138	30500	0.0283	-	-
3.9266	30600	0.0269	-	-
3.9394	30700	0.0251	-	-
3.9523	30800	0.0259	-	-
3.9651	30900	0.027	-	-
3.9779	31000	0.027	0.0235	0.5964
3.9908	31100	0.0256	-	-
4.0036	31200	0.0269	-	-
4.0164	31300	0.0233	-	-
4.0293	31400	0.0234	-	-
4.0421	31500	0.0237	-	-
4.0549	31600	0.0242	-	-
4.0678	31700	0.0232	-	-
4.0806	31800	0.0234	-	-
4.0934	31900	0.025	-	-
4.1062	32000	0.0229	0.0252	0.5908
4.1191	32100	0.0237	-	-
4.1319	32200	0.0246	-	-
4.1447	32300	0.0224	-	-
4.1576	32400	0.024	-	-
4.1704	32500	0.0238	-	-
4.1832	32600	0.0226	-	-
4.1961	32700	0.0249	-	-
4.2089	32800	0.0224	-	-
4.2217	32900	0.0226	-	-
4.2346	33000	0.0223	0.0238	0.5886
4.2474	33100	0.0237	-	-
4.2602	33200	0.025	-	-
4.2731	33300	0.0249	-	-
4.2859	33400	0.0229	-	-
4.2987	33500	0.0224	-	-
4.3116	33600	0.0237	-	-
4.3244	33700	0.0234	-	-
4.3372	33800	0.0225	-	-
4.3501	33900	0.0238	-	-
4.3629	34000	0.0245	0.0238	0.5998
4.3757	34100	0.0233	-	-
4.3886	34200	0.0241	-	-
4.4014	34300	0.0236	-	-
4.4142	34400	0.024	-	-
4.4270	34500	0.0236	-	-
4.4399	34600	0.0227	-	-
4.4527	34700	0.0214	-	-
4.4655	34800	0.0234	-	-
4.4784	34900	0.0233	-	-
4.4912	35000	0.0227	0.0229	0.5854
4.5040	35100	0.0243	-	-
4.5169	35200	0.0229	-	-
4.5297	35300	0.022	-	-
4.5425	35400	0.0235	-	-
4.5554	35500	0.0244	-	-
4.5682	35600	0.0231	-	-
4.5810	35700	0.0245	-	-
4.5939	35800	0.0224	-	-
4.6067	35900	0.0228	-	-
4.6195	36000	0.0224	0.0226	0.5926
4.6324	36100	0.0223	-	-
4.6452	36200	0.0211	-	-
4.6580	36300	0.0225	-	-
4.6709	36400	0.0236	-	-
4.6837	36500	0.0222	-	-
4.6965	36600	0.0222	-	-
4.7094	36700	0.0226	-	-
4.7222	36800	0.0233	-	-
4.7350	36900	0.0236	-	-
4.7479	37000	0.0251	0.0238	0.5930
4.7607	37100	0.0245	-	-
4.7735	37200	0.0236	-	-
4.7863	37300	0.025	-	-
4.7992	37400	0.0246	-	-
4.8120	37500	0.0235	-	-
4.8248	37600	0.0228	-	-
4.8377	37700	0.0241	-	-
4.8505	37800	0.0228	-	-
4.8633	37900	0.0211	-	-
4.8762	38000	0.0233	0.0225	0.5980
4.8890	38100	0.0231	-	-
4.9018	38200	0.0238	-	-
4.9147	38300	0.0237	-	-
4.9275	38400	0.0225	-	-
4.9403	38500	0.0237	-	-
4.9532	38600	0.0258	-	-
4.9660	38700	0.0227	-	-
4.9788	38800	0.0235	-	-
4.9917	38900	0.0234	-	-
5.0045	39000	0.0229	0.0221	0.5950
5.0173	39100	0.0206	-	-
5.0302	39200	0.0211	-	-
5.0430	39300	0.0204	-	-
5.0558	39400	0.0207	-	-
5.0687	39500	0.0196	-	-
5.0815	39600	0.0216	-	-
5.0943	39700	0.0208	-	-
5.1071	39800	0.0219	-	-
5.1200	39900	0.0218	-	-
5.1328	40000	0.0214	0.0223	0.6008
5.1456	40100	0.0199	-	-
5.1585	40200	0.0213	-	-
5.1713	40300	0.0202	-	-
5.1841	40400	0.0214	-	-
5.1970	40500	0.0205	-	-
5.2098	40600	0.0218	-	-
5.2226	40700	0.0196	-	-
5.2355	40800	0.0216	-	-
5.2483	40900	0.0206	-	-
5.2611	41000	0.0222	0.0220	0.6010
5.2740	41100	0.0207	-	-
5.2868	41200	0.0216	-	-
5.2996	41300	0.0221	-	-
5.3125	41400	0.02	-	-
5.3253	41500	0.02	-	-
5.3381	41600	0.021	-	-
5.3510	41700	0.0216	-	-
5.3638	41800	0.0203	-	-
5.3766	41900	0.0211	-	-
5.3895	42000	0.0211	0.0221	0.5950
5.4023	42100	0.021	-	-
5.4151	42200	0.0193	-	-
5.4279	42300	0.0194	-	-
5.4408	42400	0.0206	-	-
5.4536	42500	0.0202	-	-
5.4664	42600	0.0205	-	-
5.4793	42700	0.0201	-	-
5.4921	42800	0.0214	-	-
5.5049	42900	0.0201	-	-
5.5178	43000	0.0197	0.0227	0.5976
5.5306	43100	0.0222	-	-
5.5434	43200	0.0216	-	-
5.5563	43300	0.022	-	-
5.5691	43400	0.0202	-	-
5.5819	43500	0.0201	-	-
5.5948	43600	0.0193	-	-
5.6076	43700	0.0198	-	-
5.6204	43800	0.0203	-	-
5.6333	43900	0.0221	-	-
5.6461	44000	0.0217	0.0222	0.6078
5.6589	44100	0.0197	-	-
5.6718	44200	0.0195	-	-
5.6846	44300	0.0208	-	-
5.6974	44400	0.0203	-	-
5.7103	44500	0.0214	-	-
5.7231	44600	0.0216	-	-
5.7359	44700	0.0209	-	-
5.7487	44800	0.0203	-	-
5.7616	44900	0.0205	-	-
5.7744	45000	0.0209	0.0225	0.6146
5.7872	45100	0.0208	-	-
5.8001	45200	0.0193	-	-
5.8129	45300	0.0207	-	-
5.8257	45400	0.0199	-	-
5.8386	45500	0.0212	-	-
5.8514	45600	0.0209	-	-
5.8642	45700	0.0205	-	-
5.8771	45800	0.0201	-	-
5.8899	45900	0.0196	-	-
5.9027	46000	0.0206	0.0218	0.6026
5.9156	46100	0.0203	-	-
5.9284	46200	0.0216	-	-
5.9412	46300	0.0193	-	-
5.9541	46400	0.0193	-	-
5.9669	46500	0.021	-	-
5.9797	46600	0.0197	-	-
5.9926	46700	0.0206	-	-
6.0054	46800	0.02	-	-
6.0182	46900	0.0178	-	-
6.0311	47000	0.0189	0.0216	0.5982
6.0439	47100	0.0195	-	-
6.0567	47200	0.0194	-	-
6.0695	47300	0.0187	-	-
6.0824	47400	0.0178	-	-
6.0952	47500	0.0185	-	-
6.1080	47600	0.018	-	-
6.1209	47700	0.0183	-	-
6.1337	47800	0.0178	-	-
6.1465	47900	0.0181	-	-
6.1594	48000	0.0178	0.0203	0.6058
6.1722	48100	0.0179	-	-
6.1850	48200	0.0191	-	-
6.1979	48300	0.0189	-	-
6.2107	48400	0.0182	-	-
6.2235	48500	0.0195	-	-
6.2364	48600	0.0184	-	-
6.2492	48700	0.0207	-	-
6.2620	48800	0.0179	-	-
6.2749	48900	0.0189	-	-
6.2877	49000	0.0188	0.0208	0.6086
6.3005	49100	0.0174	-	-
6.3134	49200	0.0188	-	-
6.3262	49300	0.0181	-	-
6.3390	49400	0.0178	-	-
6.3519	49500	0.0181	-	-
6.3647	49600	0.0174	-	-
6.3775	49700	0.0195	-	-
6.3904	49800	0.0184	-	-
6.4032	49900	0.0172	-	-
6.4160	50000	0.0189	0.0219	0.6074
6.4288	50100	0.0188	-	-
6.4417	50200	0.0194	-	-
6.4545	50300	0.0182	-	-
6.4673	50400	0.0199	-	-
6.4802	50500	0.0189	-	-
6.4930	50600	0.0178	-	-
6.5058	50700	0.0172	-	-
6.5187	50800	0.0179	-	-
6.5315	50900	0.0178	-	-
6.5443	51000	0.0178	0.0211	0.6042
6.5572	51100	0.0176	-	-
6.5700	51200	0.0194	-	-
6.5828	51300	0.0189	-	-
6.5957	51400	0.0196	-	-
6.6085	51500	0.0178	-	-
6.6213	51600	0.0177	-	-
6.6342	51700	0.0194	-	-
6.6470	51800	0.0183	-	-
6.6598	51900	0.0213	-	-
6.6727	52000	0.0184	0.0208	0.6032
6.6855	52100	0.0182	-	-
6.6983	52200	0.0188	-	-
6.7112	52300	0.0169	-	-
6.7240	52400	0.0184	-	-
6.7368	52500	0.0198	-	-
6.7496	52600	0.018	-	-
6.7625	52700	0.0194	-	-
6.7753	52800	0.0211	-	-
6.7881	52900	0.0187	-	-
6.8010	53000	0.018	0.0197	0.6082
6.8138	53100	0.0188	-	-
6.8266	53200	0.019	-	-
6.8395	53300	0.0196	-	-
6.8523	53400	0.019	-	-
6.8651	53500	0.0198	-	-
6.8780	53600	0.0177	-	-
6.8908	53700	0.0181	-	-
6.9036	53800	0.0182	-	-
6.9165	53900	0.0189	-	-
6.9293	54000	0.0193	0.0200	0.6018
6.9421	54100	0.0182	-	-
6.9550	54200	0.0183	-	-
6.9678	54300	0.0178	-	-
6.9806	54400	0.0193	-	-
6.9935	54500	0.0186	-	-
7.0063	54600	0.017	-	-
7.0191	54700	0.0178	-	-
7.0320	54800	0.016	-	-
7.0448	54900	0.0168	-	-
7.0576	55000	0.0174	0.0198	0.6056
7.0704	55100	0.0165	-	-
7.0833	55200	0.0194	-	-
7.0961	55300	0.0186	-	-
7.1089	55400	0.0181	-	-
7.1218	55500	0.0178	-	-
7.1346	55600	0.0165	-	-
7.1474	55700	0.0174	-	-
7.1603	55800	0.0163	-	-
7.1731	55900	0.018	-	-
7.1859	56000	0.0187	0.0199	0.6034
7.1988	56100	0.0169	-	-
7.2116	56200	0.0182	-	-
7.2244	56300	0.0172	-	-
7.2373	56400	0.0176	-	-
7.2501	56500	0.0174	-	-
7.2629	56600	0.017	-	-
7.2758	56700	0.0171	-	-
7.2886	56800	0.0173	-	-
7.3014	56900	0.0173	-	-
7.3143	57000	0.0176	0.0195	0.6072
7.3271	57100	0.0166	-	-
7.3399	57200	0.0176	-	-
7.3528	57300	0.0162	-	-
7.3656	57400	0.0178	-	-
7.3784	57500	0.0164	-	-
7.3912	57600	0.0177	-	-
7.4041	57700	0.017	-	-
7.4169	57800	0.0169	-	-
7.4297	57900	0.0185	-	-
7.4426	58000	0.0165	0.0196	0.6160
7.4554	58100	0.0181	-	-
7.4682	58200	0.0172	-	-
7.4811	58300	0.0172	-	-
7.4939	58400	0.0177	-	-
7.5067	58500	0.017	-	-
7.5196	58600	0.0165	-	-
7.5324	58700	0.0178	-	-
7.5452	58800	0.0175	-	-
7.5581	58900	0.0164	-	-
7.5709	59000	0.0169	0.0194	0.6078
7.5837	59100	0.0178	-	-
7.5966	59200	0.0167	-	-
7.6094	59300	0.0173	-	-
7.6222	59400	0.0154	-	-
7.6351	59500	0.0165	-	-
7.6479	59600	0.0174	-	-
7.6607	59700	0.0166	-	-
7.6736	59800	0.0171	-	-
7.6864	59900	0.0173	-	-
7.6992	60000	0.0174	0.0192	0.6092
7.7120	60100	0.0181	-	-
7.7249	60200	0.0172	-	-
7.7377	60300	0.017	-	-
7.7505	60400	0.0172	-	-
7.7634	60500	0.0171	-	-
7.7762	60600	0.0177	-	-
7.7890	60700	0.0163	-	-
7.8019	60800	0.0163	-	-
7.8147	60900	0.017	-	-
7.8275	61000	0.0179	0.0193	0.6104
7.8404	61100	0.0161	-	-
7.8532	61200	0.0165	-	-
7.8660	61300	0.0174	-	-
7.8789	61400	0.0179	-	-
7.8917	61500	0.0175	-	-
7.9045	61600	0.0177	-	-
7.9174	61700	0.0161	-	-
7.9302	61800	0.0166	-	-
7.9430	61900	0.0172	-	-
7.9559	62000	0.0169	0.0193	0.6112
7.9687	62100	0.0162	-	-
7.9815	62200	0.0176	-	-
7.9944	62300	0.0186	-	-
8.0072	62400	0.0158	-	-
8.0200	62500	0.0154	-	-
8.0328	62600	0.0147	-	-
8.0457	62700	0.0159	-	-
8.0585	62800	0.0154	-	-
8.0713	62900	0.017	-	-
8.0842	63000	0.0165	0.0198	0.6194
8.0970	63100	0.0164	-	-
8.1098	63200	0.0162	-	-
8.1227	63300	0.0165	-	-
8.1355	63400	0.0155	-	-
8.1483	63500	0.0162	-	-
8.1612	63600	0.0156	-	-
8.1740	63700	0.015	-	-
8.1868	63800	0.0148	-	-
8.1997	63900	0.0165	-	-
8.2125	64000	0.0165	0.0199	0.6162
8.2253	64100	0.0168	-	-
8.2382	64200	0.0161	-	-
8.2510	64300	0.0166	-	-
8.2638	64400	0.0171	-	-
8.2767	64500	0.016	-	-
8.2895	64600	0.0162	-	-
8.3023	64700	0.0176	-	-
8.3152	64800	0.0154	-	-
8.3280	64900	0.0151	-	-
8.3408	65000	0.015	0.0198	0.6142
8.3537	65100	0.0161	-	-
8.3665	65200	0.0166	-	-
8.3793	65300	0.0159	-	-
8.3921	65400	0.0166	-	-
8.4050	65500	0.0162	-	-
8.4178	65600	0.0149	-	-
8.4306	65700	0.0166	-	-
8.4435	65800	0.0157	-	-
8.4563	65900	0.0158	-	-
8.4691	66000	0.0165	0.0192	0.6146
8.4820	66100	0.0159	-	-
8.4948	66200	0.016	-	-
8.5076	66300	0.0164	-	-
8.5205	66400	0.016	-	-
8.5333	66500	0.016	-	-
8.5461	66600	0.0166	-	-
8.5590	66700	0.0151	-	-
8.5718	66800	0.0156	-	-
8.5846	66900	0.0157	-	-
8.5975	67000	0.0147	0.0198	0.6152
8.6103	67100	0.0146	-	-
8.6231	67200	0.0162	-	-
8.6360	67300	0.0164	-	-
8.6488	67400	0.0149	-	-
8.6616	67500	0.0156	-	-
8.6745	67600	0.016	-	-
8.6873	67700	0.0148	-	-
8.7001	67800	0.0162	-	-
8.7129	67900	0.0159	-	-
8.7258	68000	0.0165	0.0195	0.6120
8.7386	68100	0.0152	-	-
8.7514	68200	0.0152	-	-
8.7643	68300	0.016	-	-
8.7771	68400	0.0152	-	-
8.7899	68500	0.0153	-	-
8.8028	68600	0.0147	-	-
8.8156	68700	0.016	-	-
8.8284	68800	0.0148	-	-
8.8413	68900	0.0165	-	-
8.8541	69000	0.0159	0.0191	0.6176
8.8669	69100	0.0155	-	-
8.8798	69200	0.0168	-	-
8.8926	69300	0.0169	-	-
8.9054	69400	0.0153	-	-
8.9183	69500	0.0167	-	-
8.9311	69600	0.0159	-	-
8.9439	69700	0.0168	-	-
8.9568	69800	0.015	-	-
8.9696	69900	0.0164	-	-
8.9824	70000	0.0157	0.0191	0.6168
8.9953	70100	0.0159	-	-
9.0081	70200	0.0148	-	-
9.0209	70300	0.0149	-	-
9.0337	70400	0.0143	-	-
9.0466	70500	0.0153	-	-
9.0594	70600	0.0151	-	-
9.0722	70700	0.016	-	-
9.0851	70800	0.0153	-	-
9.0979	70900	0.0158	-	-
9.1107	71000	0.0157	0.0191	0.6128
9.1236	71100	0.0159	-	-
9.1364	71200	0.0147	-	-
9.1492	71300	0.016	-	-
9.1621	71400	0.0156	-	-
9.1749	71500	0.0153	-	-
9.1877	71600	0.0149	-	-
9.2006	71700	0.0158	-	-
9.2134	71800	0.015	-	-
9.2262	71900	0.0137	-	-
9.2391	72000	0.0167	0.0190	0.6132
9.2519	72100	0.0147	-	-
9.2647	72200	0.0152	-	-
9.2776	72300	0.0148	-	-
9.2904	72400	0.0146	-	-
9.3032	72500	0.0148	-	-
9.3161	72600	0.016	-	-
9.3289	72700	0.0157	-	-
9.3417	72800	0.0149	-	-
9.3545	72900	0.0151	-	-
9.3674	73000	0.0149	0.0189	0.6162
9.3802	73100	0.0153	-	-
9.3930	73200	0.0153	-	-
9.4059	73300	0.0151	-	-
9.4187	73400	0.0149	-	-
9.4315	73500	0.0157	-	-
9.4444	73600	0.0136	-	-
9.4572	73700	0.015	-	-
9.4700	73800	0.016	-	-
9.4829	73900	0.0143	-	-
9.4957	74000	0.0155	0.0187	0.6144
9.5085	74100	0.0153	-	-
9.5214	74200	0.0153	-	-
9.5342	74300	0.014	-	-
9.5470	74400	0.0152	-	-
9.5599	74500	0.016	-	-
9.5727	74600	0.0151	-	-
9.5855	74700	0.0144	-	-
9.5984	74800	0.0152	-	-
9.6112	74900	0.0152	-	-
9.6240	75000	0.0151	0.0187	0.6140
9.6369	75100	0.0153	-	-
9.6497	75200	0.0155	-	-
9.6625	75300	0.0148	-	-
9.6753	75400	0.0145	-	-
9.6882	75500	0.0155	-	-
9.7010	75600	0.0154	-	-
9.7138	75700	0.0154	-	-
9.7267	75800	0.015	-	-
9.7395	75900	0.0157	-	-
9.7523	76000	0.0146	0.0186	0.6126
9.7652	76100	0.0148	-	-
9.7780	76200	0.0154	-	-
9.7908	76300	0.0153	-	-
9.8037	76400	0.0146	-	-
9.8165	76500	0.0144	-	-
9.8293	76600	0.015	-	-
9.8422	76700	0.0155	-	-
9.8550	76800	0.0138	-	-
9.8678	76900	0.0162	-	-
9.8807	77000	0.014	0.0187	0.6140
9.8935	77100	0.0157	-	-
9.9063	77200	0.0156	-	-
9.9192	77300	0.0155	-	-
9.9320	77400	0.015	-	-
9.9448	77500	0.0156	-	-
9.9577	77600	0.0153	-	-
9.9705	77700	0.0152	-	-
9.9833	77800	0.0159	-	-
9.9962	77900	0.0163	-	-
-1	-1	-	-	0.6154

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

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",
}

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}
}

	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