metadata

language: []
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:1793370
  - loss:CoSENTLoss
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
datasets: []
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
widget:
  - source_sentence: ek wil bietjie moderne rock hoor
    sentences:
      - request datetime
      - turn wemo on
      - query cooking
  - source_sentence: skakel af die alarm vir woensdag ses v. m.
    sentences:
      - set alarm
      - turn hue light up
      - request weather
  - source_sentence: speel my top-gegradeerde pop liedjies asseblief
    sentences:
      - greeting
      - request fact
      - request datetime
  - source_sentence: is dit warm buite
    sentences:
      - request weather
      - play music
      - request transport
  - source_sentence: maak 'n speellys van al die eminem liedjies en speel dit met skommel
    sentences:
      - search recipe
      - recommend movie
      - play music
pipeline_tag: sentence-similarity
model-index:
  - name: >-
      SentenceTransformer based on
      sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: MiniLM dev
          type: MiniLM-dev
        metrics:
          - type: pearson_cosine
            value: 0.807743120621169
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8111451989044506
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.8090992313100879
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.8112673840020295
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.8107892143621067
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.8137277702128023
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.7013144883870261
            name: Pearson Dot
          - type: spearman_dot
            value: 0.7113684320495312
            name: Spearman Dot
          - type: pearson_max
            value: 0.8107892143621067
            name: Pearson Max
          - type: spearman_max
            value: 0.8137277702128023
            name: Spearman Max

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-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.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 384 tokens
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': 128, '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})
)

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("philipp-zettl/MiniLM-amazon_massive_intent-similarity")
# Run inference
sentences = [
    "maak 'n speellys van al die eminem liedjies en speel dit met skommel",
    'play music',
    'recommend movie',
]
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

Semantic Similarity

Dataset: MiniLM-dev
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8077
spearman_cosine	0.8111
pearson_manhattan	0.8091
spearman_manhattan	0.8113
pearson_euclidean	0.8108
spearman_euclidean	0.8137
pearson_dot	0.7013
spearman_dot	0.7114
pearson_max	0.8108
spearman_max	0.8137

Training Details

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
learning_rate: 2e-05
num_train_epochs: 1
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: 32
per_device_eval_batch_size: 32
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_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}
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: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
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
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	loss	MiniLM-dev_spearman_cosine
0.0018	100	10.7509	-	-
0.0036	200	9.8726	-	-
0.0054	300	8.9837	-	-
0.0071	400	7.3162	-	-
0.0089	500	8.2842	-	-
0.0107	600	6.2254	-	-
0.0125	700	6.1004	-	-
0.0143	800	5.8583	-	-
0.0161	900	6.3118	-	-
0.0178	1000	5.7908	2.6141	0.4045
0.0196	1100	5.6907	-	-
0.0214	1200	5.6743	-	-
0.0232	1300	5.5022	-	-
0.0250	1400	5.0283	-	-
0.0268	1500	5.2936	-	-
0.0285	1600	5.2928	-	-
0.0303	1700	5.5088	-	-
0.0321	1800	5.3125	-	-
0.0339	1900	5.7931	-	-
0.0357	2000	5.5979	2.3256	0.5075
0.0375	2100	5.3222	-	-
0.0393	2200	5.268	-	-
0.0410	2300	5.264	-	-
0.0428	2400	4.9437	-	-
0.0446	2500	4.9219	-	-
0.0464	2600	4.8656	-	-
0.0482	2700	5.2733	-	-
0.0500	2800	5.0311	-	-
0.0517	2900	5.302	-	-
0.0535	3000	5.3347	2.1545	0.6496
0.0553	3100	5.1241	-	-
0.0571	3200	5.0232	-	-
0.0589	3300	4.9932	-	-
0.0607	3400	4.9651	-	-
0.0625	3500	4.5226	-	-
0.0642	3600	4.6666	-	-
0.0660	3700	4.8979	-	-
0.0678	3800	4.9139	-	-
0.0696	3900	4.9241	-	-
0.0714	4000	5.2878	2.1118	0.6948
0.0732	4100	5.0776	-	-
0.0749	4200	4.934	-	-
0.0767	4300	4.9012	-	-
0.0785	4400	4.8835	-	-
0.0803	4500	4.5886	-	-
0.0821	4600	4.7829	-	-
0.0839	4700	4.8057	-	-
0.0856	4800	4.8761	-	-
0.0874	4900	4.6787	-	-
0.0892	5000	5.313	2.1114	0.6770
0.0910	5100	5.3036	-	-
0.0928	5200	5.0731	-	-
0.0946	5300	5.0052	-	-
0.0964	5400	4.9494	-	-
0.0981	5500	4.836	-	-
0.0999	5600	4.6319	-	-
0.1017	5700	4.667	-	-
0.1035	5800	4.9578	-	-
0.1053	5900	4.9473	-	-
0.1071	6000	4.9897	3.0813	0.4424
0.1088	6100	5.1704	-	-
0.1106	6200	4.8472	-	-
0.1124	6300	4.8296	-	-
0.1142	6400	4.8287	-	-
0.1160	6500	4.6539	-	-
0.1178	6600	4.2599	-	-
0.1196	6700	4.5506	-	-
0.1213	6800	4.6585	-	-
0.1231	6900	4.7248	-	-
0.1249	7000	4.6389	3.1390	0.5199
0.1267	7100	4.8133	-	-
0.1285	7200	4.8838	-	-
0.1303	7300	4.7375	-	-
0.1320	7400	4.6357	-	-
0.1338	7500	4.7807	-	-
0.1356	7600	4.409	-	-
0.1374	7700	4.5612	-	-
0.1392	7800	4.3731	-	-
0.1410	7900	4.622	-	-
0.1427	8000	4.5574	2.6558	0.5814
0.1445	8100	4.6542	-	-
0.1463	8200	4.7831	-	-
0.1481	8300	4.6775	-	-
0.1499	8400	4.61	-	-
0.1517	8500	4.6416	-	-
0.1535	8600	4.3096	-	-
0.1552	8700	4.2629	-	-
0.1570	8800	4.5151	-	-
0.1588	8900	4.5301	-	-
0.1606	9000	4.5731	2.8939	0.5675
0.1624	9100	4.4347	-	-
0.1642	9200	4.648	-	-
0.1659	9300	4.6076	-	-
0.1677	9400	4.4229	-	-
0.1695	9500	4.4785	-	-
0.1713	9600	4.4252	-	-
0.1731	9700	4.0223	-	-
0.1749	9800	4.1593	-	-
0.1767	9900	4.2946	-	-
0.1784	10000	4.4888	2.7814	0.5852
0.1802	10100	4.3605	-	-
0.1820	10200	4.5952	-	-
0.1838	10300	4.709	-	-
0.1856	10400	4.5743	-	-
0.1874	10500	4.5539	-	-
0.1891	10600	4.4427	-	-
0.1909	10700	4.1095	-	-
0.1927	10800	4.4079	-	-
0.1945	10900	4.1667	-	-
0.1963	11000	4.2273	3.3803	0.5663
0.1981	11100	4.3333	-	-
0.1998	11200	4.5174	-	-
0.2016	11300	4.4961	-	-
0.2034	11400	4.5746	-	-
0.2052	11500	4.731	-	-
0.2070	11600	4.4485	-	-
0.2088	11700	4.4099	-	-
0.2106	11800	3.8921	-	-
0.2123	11900	4.2423	-	-
0.2141	12000	4.2641	3.0230	0.6300
0.2159	12100	4.2052	-	-
0.2177	12200	4.2757	-	-
0.2195	12300	4.8586	-	-
0.2213	12400	4.5872	-	-
0.2230	12500	4.4273	-	-
0.2248	12600	4.5728	-	-
0.2266	12700	4.4607	-	-
0.2284	12800	4.1361	-	-
0.2302	12900	4.4781	-	-
0.2320	13000	4.145	2.7088	0.6617
0.2337	13100	4.3366	-	-
0.2355	13200	4.2699	-	-
0.2373	13300	4.3397	-	-
0.2391	13400	4.6033	-	-
0.2409	13500	4.2292	-	-
0.2427	13600	4.3399	-	-
0.2445	13700	4.5222	-	-
0.2462	13800	4.2185	-	-
0.2480	13900	3.9426	-	-
0.2498	14000	4.2146	2.6014	0.6724
0.2516	14100	4.2534	-	-
0.2534	14200	4.1765	-	-
0.2552	14300	4.117	-	-
0.2569	14400	5.0908	-	-
0.2587	14500	4.488	-	-
0.2605	14600	4.4429	-	-
0.2623	14700	4.3688	-	-
0.2641	14800	4.4857	-	-
0.2659	14900	4.1763	-	-
0.2677	15000	4.4425	2.6388	0.6842
0.2694	15100	4.4277	-	-
0.2712	15200	4.3841	-	-
0.2730	15300	4.4	-	-
0.2748	15400	4.55	-	-
0.2766	15500	4.4769	-	-
0.2784	15600	4.3918	-	-
0.2801	15700	4.554	-	-
0.2819	15800	4.406	-	-
0.2837	15900	4.0593	-	-
0.2855	16000	4.3586	2.5251	0.7238
0.2873	16100	4.2308	-	-
0.2891	16200	4.469	-	-
0.2908	16300	4.2312	-	-
0.2926	16400	4.2695	-	-
0.2944	16500	4.5821	-	-
0.2962	16600	4.5623	-	-
0.2980	16700	4.1865	-	-
0.2998	16800	4.4228	-	-
0.3016	16900	4.0553	-	-
0.3033	17000	3.7183	2.6050	0.7319
0.3051	17100	4.1849	-	-
0.3069	17200	4.2975	-	-
0.3087	17300	4.4272	-	-
0.3105	17400	4.0634	-	-
0.3123	17500	4.8608	-	-
0.3140	17600	4.4146	-	-
0.3158	17700	4.2655	-	-
0.3176	17800	4.3814	-	-
0.3194	17900	4.3972	-	-
0.3212	18000	3.8868	2.4737	0.7500
0.3230	18100	4.434	-	-
0.3248	18200	4.2213	-	-
0.3265	18300	4.4632	-	-
0.3283	18400	4.4001	-	-
0.3301	18500	4.8262	-	-
0.3319	18600	4.5022	-	-
0.3337	18700	4.4148	-	-
0.3355	18800	4.2182	-	-
0.3372	18900	4.2127	-	-
0.3390	19000	4.051	2.4633	0.7575
0.3408	19100	3.655	-	-
0.3426	19200	4.2441	-	-
0.3444	19300	4.3494	-	-
0.3462	19400	4.1824	-	-
0.3479	19500	4.3528	-	-
0.3497	19600	5.6073	-	-
0.3515	19700	4.8231	-	-
0.3533	19800	4.5816	-	-
0.3551	19900	4.5812	-	-
0.3569	20000	4.637	2.1229	0.7945
0.3587	20100	4.2619	-	-
0.3604	20200	4.5645	-	-
0.3622	20300	4.7248	-	-
0.3640	20400	4.5665	-	-
0.3658	20500	4.5628	-	-
0.3676	20600	4.8494	-	-
0.3694	20700	4.4338	-	-
0.3711	20800	4.3256	-	-
0.3729	20900	4.4388	-	-
0.3747	21000	4.158	2.3475	0.7732
0.3765	21100	3.962	-	-
0.3783	21200	3.931	-	-
0.3801	21300	4.0345	-	-
0.3818	21400	4.319	-	-
0.3836	21500	4.1329	-	-
0.3854	21600	4.245	-	-
0.3872	21700	4.518	-	-
0.3890	21800	4.4653	-	-
0.3908	21900	4.2777	-	-
0.3926	22000	4.3358	2.1933	0.7845
0.3943	22100	4.2291	-	-
0.3961	22200	3.8067	-	-
0.3979	22300	4.2039	-	-
0.3997	22400	4.0104	-	-
0.4015	22500	4.2346	-	-
0.4033	22600	4.0056	-	-
0.4050	22700	5.6038	-	-
0.4068	22800	5.1185	-	-
0.4086	22900	4.924	-	-
0.4104	23000	4.7841	1.9839	0.7956
0.4122	23100	4.7953	-	-
0.4140	23200	4.4229	-	-
0.4158	23300	4.6432	-	-
0.4175	23400	4.5284	-	-
0.4193	23500	4.7215	-	-
0.4211	23600	4.7432	-	-
0.4229	23700	5.0136	-	-
0.4247	23800	4.7958	-	-
0.4265	23900	4.6827	-	-
0.4282	24000	4.6665	1.9663	0.7870
0.4300	24100	4.5074	-	-
0.4318	24200	4.4189	-	-
0.4336	24300	4.4586	-	-
0.4354	24400	4.6421	-	-
0.4372	24500	4.4281	-	-
0.4389	24600	4.5153	-	-
0.4407	24700	4.9942	-	-
0.4425	24800	5.11	-	-
0.4443	24900	4.7071	-	-
0.4461	25000	4.6257	1.9461	0.7935
0.4479	25100	4.6576	-	-
0.4497	25200	4.6103	-	-
0.4514	25300	4.2066	-	-
0.4532	25400	4.6869	-	-
0.4550	25500	4.7575	-	-
0.4568	25600	4.6081	-	-
0.4586	25700	4.8144	-	-
0.4604	25800	5.2007	-	-
0.4621	25900	4.8367	-	-
0.4639	26000	4.5258	1.9131	0.7993
0.4657	26100	4.4784	-	-
0.4675	26200	4.5568	-	-
0.4693	26300	4.2591	-	-
0.4711	26400	4.4521	-	-
0.4729	26500	4.4041	-	-
0.4746	26600	4.4926	-	-
0.4764	26700	4.1686	-	-
0.4782	26800	4.6294	-	-
0.4800	26900	4.6889	-	-
0.4818	27000	4.5765	1.9539	0.7961
0.4836	27100	4.3427	-	-
0.4853	27200	4.5275	-	-
0.4871	27300	4.4186	-	-
0.4889	27400	4.0163	-	-
0.4907	27500	4.3204	-	-
0.4925	27600	4.179	-	-
0.4943	27700	4.3838	-	-
0.4960	27800	4.2631	-	-
0.4978	27900	4.7177	-	-
0.4996	28000	4.5161	2.0116	0.7935
0.5014	28100	4.2861	-	-
0.5032	28200	4.4123	-	-
0.5050	28300	4.293	-	-
0.5068	28400	4.2346	-	-
0.5085	28500	4.3355	-	-
0.5103	28600	4.4616	-	-
0.5121	28700	4.2409	-	-
0.5139	28800	4.2398	-	-
0.5157	28900	4.7412	-	-
0.5175	29000	4.5044	2.1008	0.7859
0.5192	29100	4.4556	-	-
0.5210	29200	4.2938	-	-
0.5228	29300	4.4962	-	-
0.5246	29400	4.477	-	-
0.5264	29500	4.2602	-	-
0.5282	29600	4.4231	-	-
0.5300	29700	4.2165	-	-
0.5317	29800	4.3729	-	-
0.5335	29900	4.2414	-	-
0.5353	30000	4.9937	2.0884	0.7702
0.5371	30100	4.5737	-	-
0.5389	30200	4.4517	-	-
0.5407	30300	4.4178	-	-
0.5424	30400	4.3514	-	-
0.5442	30500	3.9723	-	-
0.5460	30600	4.3707	-	-
0.5478	30700	4.2235	-	-
0.5496	30800	4.4278	-	-
0.5514	30900	4.2914	-	-
0.5531	31000	4.5636	2.3277	0.7454
0.5549	31100	4.4889	-	-
0.5567	31200	4.3211	-	-
0.5585	31300	4.404	-	-
0.5603	31400	4.2117	-	-
0.5621	31500	4.1126	-	-
0.5639	31600	4.1737	-	-
0.5656	31700	4.203	-	-
0.5674	31800	4.1093	-	-
0.5692	31900	4.0702	-	-
0.5710	32000	4.4189	2.6265	0.7375
0.5728	32100	4.9817	-	-
0.5746	32200	4.4736	-	-
0.5763	32300	4.348	-	-
0.5781	32400	4.5404	-	-
0.5799	32500	4.2987	-	-
0.5817	32600	4.0725	-	-
0.5835	32700	4.5469	-	-
0.5853	32800	4.4367	-	-
0.5870	32900	4.3369	-	-
0.5888	33000	4.2292	2.5687	0.7213
0.5906	33100	4.7929	-	-
0.5924	33200	4.4123	-	-
0.5942	33300	4.1699	-	-
0.5960	33400	4.4021	-	-
0.5978	33500	4.5257	-	-
0.5995	33600	3.7222	-	-
0.6013	33700	4.0746	-	-
0.6031	33800	4.1399	-	-
0.6049	33900	3.9957	-	-
0.6067	34000	4.093	2.4645	0.7524
0.6085	34100	4.2929	-	-
0.6102	34200	4.4765	-	-
0.6120	34300	4.3871	-	-
0.6138	34400	4.385	-	-
0.6156	34500	4.1455	-	-
0.6174	34600	3.7689	-	-
0.6192	34700	3.6574	-	-
0.6210	34800	4.2426	-	-
0.6227	34900	4.293	-	-
0.6245	35000	4.1368	2.4370	0.7765
0.6263	35100	3.6174	-	-
0.6281	35200	4.7763	-	-
0.6299	35300	4.3121	-	-
0.6317	35400	4.1886	-	-
0.6334	35500	4.3538	-	-
0.6352	35600	4.0285	-	-
0.6370	35700	3.4691	-	-
0.6388	35800	4.2732	-	-
0.6406	35900	4.2052	-	-
0.6424	36000	4.0452	2.4680	0.7732
0.6441	36100	3.9032	-	-
0.6459	36200	4.2608	-	-
0.6477	36300	4.262	-	-
0.6495	36400	4.1138	-	-
0.6513	36500	4.248	-	-
0.6531	36600	4.1163	-	-
0.6549	36700	3.6375	-	-
0.6566	36800	4.0768	-	-
0.6584	36900	4.0268	-	-
0.6602	37000	4.0129	2.6361	0.7702
0.6620	37100	3.7976	-	-
0.6638	37200	4.2518	-	-
0.6656	37300	4.5011	-	-
0.6673	37400	4.4488	-	-
0.6691	37500	3.9798	-	-
0.6709	37600	4.027	-	-
0.6727	37700	4.0342	-	-
0.6745	37800	3.8229	-	-
0.6763	37900	4.0573	-	-
0.6781	38000	4.1739	2.4511	0.7935
0.6798	38100	4.57	-	-
0.6816	38200	3.9108	-	-
0.6834	38300	4.3569	-	-
0.6852	38400	4.3775	-	-
0.6870	38500	4.2887	-	-
0.6888	38600	4.144	-	-
0.6905	38700	4.5112	-	-
0.6923	38800	3.5093	-	-
0.6941	38900	3.9626	-	-
0.6959	39000	4.024	2.4241	0.7868
0.6977	39100	4.0671	-	-
0.6995	39200	3.9545	-	-
0.7012	39300	4.0036	-	-
0.7030	39400	4.3796	-	-
0.7048	39500	4.2912	-	-
0.7066	39600	4.1181	-	-
0.7084	39700	4.1437	-	-
0.7102	39800	3.8734	-	-
0.7120	39900	3.7678	-	-
0.7137	40000	4.2327	2.3937	0.7956
0.7155	40100	3.8276	-	-
0.7173	40200	4.2885	-	-
0.7191	40300	4.019	-	-
0.7209	40400	4.6898	-	-
0.7227	40500	4.2398	-	-
0.7244	40600	4.317	-	-
0.7262	40700	4.2543	-	-
0.7280	40800	4.1048	-	-
0.7298	40900	3.4243	-	-
0.7316	41000	4.0587	2.2848	0.8035
0.7334	41100	4.2112	-	-
0.7351	41200	4.0331	-	-
0.7369	41300	4.2361	-	-
0.7387	41400	4.3818	-	-
0.7405	41500	4.1311	-	-
0.7423	41600	4.0607	-	-
0.7441	41700	4.1277	-	-
0.7459	41800	3.8844	-	-
0.7476	41900	3.6138	-	-
0.7494	42000	3.7973	2.4197	0.8045
0.7512	42100	4.0854	-	-
0.7530	42200	4.0926	-	-
0.7548	42300	3.9821	-	-
0.7566	42400	4.5564	-	-
0.7583	42500	6.1707	-	-
0.7601	42600	5.4598	-	-
0.7619	42700	5.2202	-	-
0.7637	42800	5.1402	-	-
0.7655	42900	4.8446	-	-
0.7673	43000	4.5341	1.9710	0.8181
0.7691	43100	5.0068	-	-
0.7708	43200	5.0099	-	-
0.7726	43300	4.7986	-	-
0.7744	43400	5.0468	-	-
0.7762	43500	5.135	-	-
0.7780	43600	4.8018	-	-
0.7798	43700	4.6291	-	-
0.7815	43800	4.6119	-	-
0.7833	43900	4.5318	-	-
0.7851	44000	3.9703	1.9790	0.8211
0.7869	44100	4.461	-	-
0.7887	44200	4.5536	-	-
0.7905	44300	4.411	-	-
0.7922	44400	4.5796	-	-
0.7940	44500	4.7385	-	-
0.7958	44600	4.6635	-	-
0.7976	44700	4.4808	-	-
0.7994	44800	4.5565	-	-
0.8012	44900	4.4707	-	-
0.8030	45000	3.9981	1.9823	0.8197
0.8047	45100	4.119	-	-
0.8065	45200	4.4209	-	-
0.8083	45300	4.3268	-	-
0.8101	45400	4.2979	-	-
0.8119	45500	4.413	-	-
0.8137	45600	4.3317	-	-
0.8154	45700	4.3683	-	-
0.8172	45800	4.0769	-	-
0.8190	45900	4.304	-	-
0.8208	46000	4.0985	2.0490	0.8183
0.8226	46100	3.8719	-	-
0.8244	46200	4.1843	-	-
0.8262	46300	4.2131	-	-
0.8279	46400	4.3327	-	-
0.8297	46500	3.8533	-	-
0.8315	46600	5.2854	-	-
0.8333	46700	5.2465	-	-
0.8351	46800	5.0221	-	-
0.8369	46900	4.9466	-	-
0.8386	47000	5.0361	1.8252	0.8360
0.8404	47100	4.3676	-	-
0.8422	47200	4.619	-	-
0.8440	47300	4.6412	-	-
0.8458	47400	4.7874	-	-
0.8476	47500	4.663	-	-
0.8493	47600	4.7068	-	-
0.8511	47700	4.5889	-	-
0.8529	47800	4.3468	-	-
0.8547	47900	4.4393	-	-
0.8565	48000	4.5488	1.9117	0.8176
0.8583	48100	4.0933	-	-
0.8601	48200	3.7754	-	-
0.8618	48300	4.1346	-	-
0.8636	48400	4.402	-	-
0.8654	48500	4.0163	-	-
0.8672	48600	4.3405	-	-
0.8690	48700	4.7694	-	-
0.8708	48800	4.4457	-	-
0.8725	48900	4.3679	-	-
0.8743	49000	4.3283	1.9392	0.8251
0.8761	49100	4.6855	-	-
0.8779	49200	3.881	-	-
0.8797	49300	4.1392	-	-
0.8815	49400	4.4343	-	-
0.8833	49500	4.4822	-	-
0.8850	49600	4.3977	-	-
0.8868	49700	4.5944	-	-
0.8886	49800	4.4176	-	-
0.8904	49900	4.5269	-	-
0.8922	50000	4.4267	1.8965	0.8206
0.8940	50100	4.5109	-	-
0.8957	50200	4.1775	-	-
0.8975	50300	4.3453	-	-
0.8993	50400	4.5443	-	-
0.9011	50500	4.226	-	-
0.9029	50600	4.3296	-	-
0.9047	50700	4.1968	-	-
0.9064	50800	4.2206	-	-
0.9082	50900	4.2299	-	-
0.9100	51000	4.0471	2.0479	0.8146
0.9118	51100	4.0832	-	-
0.9136	51200	3.7516	-	-
0.9154	51300	4.0545	-	-
0.9172	51400	4.1281	-	-
0.9189	51500	4.2336	-	-
0.9207	51600	4.2511	-	-
0.9225	51700	4.2588	-	-
0.9243	51800	4.0719	-	-
0.9261	51900	4.1847	-	-
0.9279	52000	4.1445	2.1419	0.8128
0.9296	52100	3.9735	-	-
0.9314	52200	3.8635	-	-
0.9332	52300	4.1738	-	-
0.9350	52400	4.07	-	-
0.9368	52500	4.1008	-	-
0.9386	52600	3.9628	-	-
0.9403	52700	4.2895	-	-
0.9421	52800	4.3393	-	-
0.9439	52900	2.8535	-	-
0.9457	53000	2.5506	2.1743	0.8116
0.9475	53100	2.1566	-	-
0.9493	53200	2.0386	-	-
0.9511	53300	1.8535	-	-
0.9528	53400	1.8561	-	-
0.9546	53500	1.3213	-	-
0.9564	53600	1.0904	-	-
0.9582	53700	1.2266	-	-
0.9600	53800	0.9386	-	-
0.9618	53900	0.8379	-	-
0.9635	54000	0.9314	2.3331	0.8071
0.9653	54100	1.1145	-	-
0.9671	54200	1.4435	-	-
0.9689	54300	1.3226	-	-
0.9707	54400	0.6677	-	-
0.9725	54500	0.7357	-	-
0.9743	54600	0.6854	-	-
0.9760	54700	0.8408	-	-
0.9778	54800	0.6291	-	-
0.9796	54900	0.8203	-	-
0.9814	55000	1.6263	2.4720	0.8104
0.9832	55100	0.95	-	-
0.9850	55200	0.6462	-	-
0.9867	55300	1.2467	-	-
0.9885	55400	1.4926	-	-
0.9903	55500	1.9608	-	-
0.9921	55600	1.6415	-	-
0.9939	55700	1.3258	-	-
0.9957	55800	1.2157	-	-
0.9974	55900	1.2391	-	-
0.9992	56000	1.3474	2.5008	0.8111

Framework Versions

Python: 3.10.14
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.1+cu121
Accelerate: 0.33.0
Datasets: 2.21.0
Tokenizers: 0.19.1

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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}