metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:21210762
  - loss:MSELoss
base_model: sentence-transformers/all-MiniLM-L6-v2
widget:
  - source_sentence: >-
      A Message Authentication Code is a protection against data being altered
      in transit by an attacker who has the ability to read the data in
      real-time.
    sentences:
      - 彼との仕事はめちゃくちゃクールだった。
      - もう何十台も通り過ぎた。
      - メッセージ認証コードは、データをリアルタイムで読み取る能力を持つ攻撃者による転送中のデータの改ざんから保護します。
  - source_sentence: >-
      “We have the best entertainers from New York, Hollywood and Las Vegas
      perform here.”
    sentences:
      - 「ニューヨークやハリウッド、ラスベガスからの素晴らしい芸人たちがここでショーをするんです」
      - 現在は様々なサプリが販売されています。
      - ◆　トルコの友人たちへの手紙
  - source_sentence: A correction was made on November 24th.
    sentences:
      - 私たちが抱え込んでいたトラウマ。
      - 11月24日に訂正いたしました。
      - なぜ憲法を学ぶのでしょうか。
  - source_sentence: We need more teachers like him nowadays.
    sentences:
      - そういうことが、いまの教員には、もっと必要でしょうね。
      - >-
        This may include, but is not limited to, investigating and intercepting
        payments into and out of your account(s) (particularly in the case of
        international transfers of funds) and investigating the source of or
        intended recipient of funds.
        ⇒　少なくとも（違法行為が疑われる）名義人の口座の出入金（特に国際的な送金）について調査し、かつそれらを差し止めたうえで、入金についてはその送金元、出金についてはその受取り人を調査します。
      - 私がそこで授業をしているのである。
  - source_sentence: >-
      Nano-structure A nanostructure is an intermediate size between molecular
      and microscopic (micrometer-sized) structures.
    sentences:
      - 魔術のお話に戻りましょう。
      - 引き続きワシントンより。
      - ナノ構造は、分子構造と微視的（マイクロメートルサイズ）構造との間の中間サイズの対象である。
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: JSTS
          type: JSTS
        metrics:
          - type: pearson_cosine
            value: 0.823463331969533
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.7815308480362135
            name: Spearman Cosine
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: stsb multi mt en
          type: stsb_multi_mt-en
        metrics:
          - type: pearson_cosine
            value: 0.8362828278686943
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8564038929573722
            name: Spearman Cosine

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

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-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/all-MiniLM-L6-v2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 384 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': 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})
  (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("sentence_transformers_model_id")
# Run inference
sentences = [
    'Nano-structure A nanostructure is an intermediate size between molecular and microscopic (micrometer-sized) structures.',
    'ナノ構造は、分子構造と微視的（マイクロメートルサイズ）構造との間の中間サイズの対象である。',
    '魔術のお話に戻りましょう。',
]
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

Datasets: JSTS and stsb_multi_mt-en
Evaluated with EmbeddingSimilarityEvaluator

Metric	JSTS	stsb_multi_mt-en
pearson_cosine	0.8235	0.8363
spearman_cosine	0.7815	0.8564

Training Details

Training Dataset

Unnamed Dataset

Size: 21,210,762 training samples
Columns: english, non_english, and label
Approximate statistics based on the first 1000 samples:
english non_english label
type string string list
details
min: 4 tokens
mean: 16.46 tokens
max: 92 tokens

min: 4 tokens
mean: 21.99 tokens
max: 128 tokens

size: 384 elements

	english	non_english	label
type	string	string	list
details	min: 4 tokens mean: 16.46 tokens max: 92 tokens	min: 4 tokens mean: 21.99 tokens max: 128 tokens	size: 384 elements

Samples:

english	non_english	label
`We live the life of the project.`	`プロジェクトの生命を左右する。`	`[0.009600206278264523, 0.058811139315366745, 0.023707984015345573, -0.021880649030208588, 0.068634033203125, ...]`
`Hold on here, Mr. Budget Director.`	`ここいろ編集長`	`[-0.04940887540578842, -0.013437069952487946, 0.024199623614549637, -0.02371774986386299, 0.06858911365270615, ...]`
`So yes, biology has all the attributes of a transportation genius today.`	`そうです生物は今日話した最高の交通にある特性を全て持ち合わせています`	`[0.031787291169166565, 0.011292539536952972, 0.03621761128306389, -0.04237872734665871, -0.030112963169813156, ...]`

Loss: MSELoss

Evaluation Dataset

Unnamed Dataset

Size: 214,251 evaluation samples
Columns: english, non_english, and label
Approximate statistics based on the first 1000 samples:
english non_english label
type string string list
details
min: 4 tokens
mean: 16.24 tokens
max: 88 tokens

min: 4 tokens
mean: 22.18 tokens
max: 128 tokens

size: 384 elements

	english	non_english	label
type	string	string	list
details	min: 4 tokens mean: 16.24 tokens max: 88 tokens	min: 4 tokens mean: 22.18 tokens max: 128 tokens	size: 384 elements

Samples:

english	non_english	label
`Then the next step was the social bookmarking.`	`次のカテゴリはソーシャルブックマークです。`	`[-0.040418993681669235, 0.019537044689059258, -0.014964035712182522, -0.06385297328233719, 0.00023657231940887868, ...]`
`Ooh! Scary word! Ahh!`	`なんと恐ろしい言葉！`	`[0.023886308073997498, -0.04336044192314148, -0.057255394756793976, 0.05142980441451073, 0.06282227486371994, ...]`
`Usually Ebates offers 1.`	`通常提示スプレッド*1`	`[0.00018616259330883622, -0.01999301090836525, 0.049356017261743546, 0.002617522142827511, -0.0540102981030941, ...]`

Loss: MSELoss

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 512
per_device_eval_batch_size: 512
gradient_accumulation_steps: 2
learning_rate: 0.0003
num_train_epochs: 8
warmup_ratio: 0.15
bf16: True
dataloader_num_workers: 8

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 512
per_device_eval_batch_size: 512
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 2
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 0.0003
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 8
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.15
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: True
fp16: False
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: 8
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
tp_size: 0
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss	JSTS_spearman_cosine	stsb_multi_mt-en_spearman_cosine
0.0241	500	0.0064	-	-	-
0.0483	1000	0.0045	-	-	-
0.0724	1500	0.0038	-	-	-
0.0966	2000	0.0035	0.0016	0.3008	0.2821
0.1207	2500	0.0033	-	-	-
0.1448	3000	0.0031	-	-	-
0.1690	3500	0.0029	-	-	-
0.1931	4000	0.0028	0.0013	0.4989	0.4681
0.2172	4500	0.0026	-	-	-
0.2414	5000	0.0025	-	-	-
0.2655	5500	0.0023	-	-	-
0.2897	6000	0.0022	0.0010	0.6554	0.6567
0.3138	6500	0.0021	-	-	-
0.3379	7000	0.002	-	-	-
0.3621	7500	0.0019	-	-	-
0.3862	8000	0.0018	0.0008	0.7038	0.7328
0.4104	8500	0.0017	-	-	-
0.4345	9000	0.0017	-	-	-
0.4586	9500	0.0016	-	-	-
0.4828	10000	0.0016	0.0007	0.7420	0.7662
0.5069	10500	0.0015	-	-	-
0.5310	11000	0.0015	-	-	-
0.5552	11500	0.0014	-	-	-
0.5793	12000	0.0014	0.0006	0.7559	0.7929
0.6035	12500	0.0014	-	-	-
0.6276	13000	0.0014	-	-	-
0.6517	13500	0.0013	-	-	-
0.6759	14000	0.0013	0.0006	0.7625	0.8056
0.7000	14500	0.0013	-	-	-
0.7241	15000	0.0013	-	-	-
0.7483	15500	0.0012	-	-	-
0.7724	16000	0.0012	0.0006	0.7652	0.8150
0.7966	16500	0.0012	-	-	-
0.8207	17000	0.0012	-	-	-
0.8448	17500	0.0012	-	-	-
0.8690	18000	0.0012	0.0005	0.7679	0.8209
0.8931	18500	0.0012	-	-	-
0.9173	19000	0.0011	-	-	-
0.9414	19500	0.0011	-	-	-
0.9655	20000	0.0011	0.0005	0.7727	0.8269
0.9897	20500	0.0011	-	-	-
1.0138	21000	0.0011	-	-	-
1.0379	21500	0.0011	-	-	-
1.0621	22000	0.0011	0.0005	0.7682	0.8319
1.0862	22500	0.0011	-	-	-
1.1104	23000	0.0011	-	-	-
1.1345	23500	0.0011	-	-	-
1.1586	24000	0.0011	0.0005	0.7718	0.8372
1.1828	24500	0.0011	-	-	-
1.2069	25000	0.0011	-	-	-
1.2311	25500	0.0011	-	-	-
1.2552	26000	0.001	0.0005	0.7751	0.8408
1.2793	26500	0.001	-	-	-
1.3035	27000	0.001	-	-	-
1.3276	27500	0.001	-	-	-
1.3517	28000	0.001	0.0005	0.7703	0.8437
1.3759	28500	0.001	-	-	-
1.4000	29000	0.001	-	-	-
1.4242	29500	0.001	-	-	-
1.4483	30000	0.001	0.0005	0.7730	0.8439
1.4724	30500	0.001	-	-	-
1.4966	31000	0.001	-	-	-
1.5207	31500	0.001	-	-	-
1.5448	32000	0.001	0.0005	0.7719	0.8456
1.5690	32500	0.001	-	-	-
1.5931	33000	0.001	-	-	-
1.6173	33500	0.001	-	-	-
1.6414	34000	0.001	0.0005	0.7719	0.8449
1.6655	34500	0.001	-	-	-
1.6897	35000	0.001	-	-	-
1.7138	35500	0.001	-	-	-
1.7380	36000	0.001	0.0004	0.7717	0.8455
1.7621	36500	0.001	-	-	-
1.7862	37000	0.001	-	-	-
1.8104	37500	0.001	-	-	-
1.8345	38000	0.001	0.0004	0.7714	0.8488
1.8586	38500	0.001	-	-	-
1.8828	39000	0.001	-	-	-
1.9069	39500	0.001	-	-	-
1.9311	40000	0.001	0.0004	0.7753	0.8474
1.9552	40500	0.001	-	-	-
1.9793	41000	0.001	-	-	-
2.0035	41500	0.001	-	-	-
2.0276	42000	0.001	0.0004	0.7708	0.8479
2.0518	42500	0.001	-	-	-
2.0759	43000	0.001	-	-	-
2.1000	43500	0.001	-	-	-
2.1242	44000	0.001	0.0004	0.7703	0.8505
2.1483	44500	0.001	-	-	-
2.1724	45000	0.0009	-	-	-
2.1966	45500	0.0009	-	-	-
2.2207	46000	0.0009	0.0004	0.7752	0.8525
2.2449	46500	0.0009	-	-	-
2.2690	47000	0.0009	-	-	-
2.2931	47500	0.0009	-	-	-
2.3173	48000	0.0009	0.0004	0.7734	0.8518
2.3414	48500	0.0009	-	-	-
2.3655	49000	0.0009	-	-	-
2.3897	49500	0.0009	-	-	-
2.4138	50000	0.0009	0.0004	0.7725	0.8512
2.4380	50500	0.0009	-	-	-
2.4621	51000	0.0009	-	-	-
2.4862	51500	0.0009	-	-	-
2.5104	52000	0.0009	0.0004	0.7709	0.8535
2.5345	52500	0.0009	-	-	-
2.5587	53000	0.0009	-	-	-
2.5828	53500	0.0009	-	-	-
2.6069	54000	0.0009	0.0004	0.7751	0.8519
2.6311	54500	0.0009	-	-	-
2.6552	55000	0.0009	-	-	-
2.6793	55500	0.0009	-	-	-
2.7035	56000	0.0009	0.0004	0.7770	0.8500
2.7276	56500	0.0009	-	-	-
2.7518	57000	0.0009	-	-	-
2.7759	57500	0.0009	-	-	-
2.8000	58000	0.0009	0.0004	0.7756	0.8514
2.8242	58500	0.0009	-	-	-
2.8483	59000	0.0009	-	-	-
2.8725	59500	0.0009	-	-	-
2.8966	60000	0.0009	0.0004	0.7791	0.8541
2.9207	60500	0.0009	-	-	-
2.9449	61000	0.0009	-	-	-
2.9690	61500	0.0009	-	-	-
2.9931	62000	0.0009	0.0004	0.7759	0.8539
3.0173	62500	0.0009	-	-	-
3.0414	63000	0.0009	-	-	-
3.0656	63500	0.0009	-	-	-
3.0897	64000	0.0009	0.0004	0.7770	0.8526
3.1138	64500	0.0009	-	-	-
3.1380	65000	0.0009	-	-	-
3.1621	65500	0.0009	-	-	-
3.1863	66000	0.0009	0.0004	0.7762	0.8531
3.2104	66500	0.0009	-	-	-
3.2345	67000	0.0009	-	-	-
3.2587	67500	0.0009	-	-	-
3.2828	68000	0.0009	0.0004	0.7771	0.8515
3.3069	68500	0.0009	-	-	-
3.3311	69000	0.0009	-	-	-
3.3552	69500	0.0009	-	-	-
3.3794	70000	0.0009	0.0004	0.7757	0.8530
3.4035	70500	0.0009	-	-	-
3.4276	71000	0.0009	-	-	-
3.4518	71500	0.0009	-	-	-
3.4759	72000	0.0009	0.0004	0.7776	0.8532
3.5000	72500	0.0009	-	-	-
3.5242	73000	0.0009	-	-	-
3.5483	73500	0.0009	-	-	-
3.5725	74000	0.0009	0.0004	0.7776	0.8542
3.5966	74500	0.0009	-	-	-
3.6207	75000	0.0009	-	-	-
3.6449	75500	0.0009	-	-	-
3.6690	76000	0.0009	0.0004	0.7803	0.8539
3.6932	76500	0.0009	-	-	-
3.7173	77000	0.0009	-	-	-
3.7414	77500	0.0009	-	-	-
3.7656	78000	0.0009	0.0004	0.7778	0.8537
3.7897	78500	0.0009	-	-	-
3.8138	79000	0.0009	-	-	-
3.8380	79500	0.0009	-	-	-
3.8621	80000	0.0009	0.0004	0.7800	0.8539
3.8863	80500	0.0009	-	-	-
3.9104	81000	0.0009	-	-	-
3.9345	81500	0.0009	-	-	-
3.9587	82000	0.0009	0.0004	0.7797	0.8542
3.9828	82500	0.0009	-	-	-
4.0070	83000	0.0009	-	-	-
4.0311	83500	0.0009	-	-	-
4.0552	84000	0.0009	0.0004	0.7808	0.8547
4.0794	84500	0.0009	-	-	-
4.1035	85000	0.0009	-	-	-
4.1276	85500	0.0009	-	-	-
4.1518	86000	0.0009	0.0004	0.7778	0.8545
4.1759	86500	0.0009	-	-	-
4.2001	87000	0.0009	-	-	-
4.2242	87500	0.0009	-	-	-
4.2483	88000	0.0009	0.0004	0.7815	0.8555
4.2725	88500	0.0009	-	-	-
4.2966	89000	0.0009	-	-	-
4.3207	89500	0.0009	-	-	-
4.3449	90000	0.0009	0.0004	0.7797	0.8534
4.3690	90500	0.0009	-	-	-
4.3932	91000	0.0009	-	-	-
4.4173	91500	0.0009	-	-	-
4.4414	92000	0.0009	0.0004	0.7823	0.8547
4.4656	92500	0.0009	-	-	-
4.4897	93000	0.0009	-	-	-
4.5139	93500	0.0009	-	-	-
4.5380	94000	0.0009	0.0004	0.7783	0.8535
4.5621	94500	0.0009	-	-	-
4.5863	95000	0.0009	-	-	-
4.6104	95500	0.0009	-	-	-
4.6345	96000	0.0009	0.0004	0.7811	0.8550
4.6587	96500	0.0009	-	-	-
4.6828	97000	0.0009	-	-	-
4.7070	97500	0.0009	-	-	-
4.7311	98000	0.0009	0.0004	0.7801	0.8540
4.7552	98500	0.0009	-	-	-
4.7794	99000	0.0009	-	-	-
4.8035	99500	0.0009	-	-	-
4.8277	100000	0.0009	0.0004	0.7811	0.8544
4.8518	100500	0.0009	-	-	-
4.8759	101000	0.0009	-	-	-
4.9001	101500	0.0009	-	-	-
4.9242	102000	0.0009	0.0004	0.7805	0.8548
4.9483	102500	0.0009	-	-	-
4.9725	103000	0.0009	-	-	-
4.9966	103500	0.0009	-	-	-
5.0208	104000	0.0009	0.0004	0.7797	0.8534
5.0449	104500	0.0009	-	-	-
5.0690	105000	0.0009	-	-	-
5.0932	105500	0.0009	-	-	-
5.1173	106000	0.0009	0.0004	0.7821	0.8555
5.1415	106500	0.0009	-	-	-
5.1656	107000	0.0009	-	-	-
5.1897	107500	0.0009	-	-	-
5.2139	108000	0.0009	0.0004	0.7816	0.8558
5.2380	108500	0.0009	-	-	-
5.2621	109000	0.0009	-	-	-
5.2863	109500	0.0009	-	-	-
5.3104	110000	0.0009	0.0004	0.7804	0.8556
5.3346	110500	0.0009	-	-	-
5.3587	111000	0.0009	-	-	-
5.3828	111500	0.0009	-	-	-
5.4070	112000	0.0009	0.0004	0.7813	0.8548
5.4311	112500	0.0009	-	-	-
5.4552	113000	0.0009	-	-	-
5.4794	113500	0.0009	-	-	-
5.5035	114000	0.0009	0.0004	0.7823	0.8548
5.5277	114500	0.0009	-	-	-
5.5518	115000	0.0009	-	-	-
5.5759	115500	0.0009	-	-	-
5.6001	116000	0.0009	0.0004	0.7809	0.8551
5.6242	116500	0.0009	-	-	-
5.6484	117000	0.0009	-	-	-
5.6725	117500	0.0009	-	-	-
5.6966	118000	0.0009	0.0004	0.7833	0.8557
5.7208	118500	0.0009	-	-	-
5.7449	119000	0.0009	-	-	-
5.7690	119500	0.0009	-	-	-
5.7932	120000	0.0009	0.0004	0.7842	0.8551
5.8173	120500	0.0009	-	-	-
5.8415	121000	0.0009	-	-	-
5.8656	121500	0.0009	-	-	-
5.8897	122000	0.0009	0.0004	0.7817	0.8563
5.9139	122500	0.0009	-	-	-
5.9380	123000	0.0009	-	-	-
5.9622	123500	0.0009	-	-	-
5.9863	124000	0.0009	0.0004	0.7812	0.8559
6.0104	124500	0.0009	-	-	-
6.0346	125000	0.0009	-	-	-
6.0587	125500	0.0009	-	-	-
6.0828	126000	0.0009	0.0004	0.7821	0.8558
6.1070	126500	0.0009	-	-	-
6.1311	127000	0.0009	-	-	-
6.1553	127500	0.0009	-	-	-
6.1794	128000	0.0009	0.0004	0.7829	0.8548
6.2035	128500	0.0009	-	-	-
6.2277	129000	0.0009	-	-	-
6.2518	129500	0.0009	-	-	-
6.2759	130000	0.0009	0.0004	0.7805	0.8549
6.3001	130500	0.0009	-	-	-
6.3242	131000	0.0009	-	-	-
6.3484	131500	0.0009	-	-	-
6.3725	132000	0.0009	0.0004	0.7807	0.8563
6.3966	132500	0.0009	-	-	-
6.4208	133000	0.0009	-	-	-
6.4449	133500	0.0009	-	-	-
6.4691	134000	0.0009	0.0004	0.7829	0.8555
6.4932	134500	0.0009	-	-	-
6.5173	135000	0.0009	-	-	-
6.5415	135500	0.0009	-	-	-
6.5656	136000	0.0009	0.0004	0.7819	0.8550
6.5897	136500	0.0009	-	-	-
6.6139	137000	0.0009	-	-	-
6.6380	137500	0.0009	-	-	-
6.6622	138000	0.0009	0.0004	0.7800	0.8548
6.6863	138500	0.0009	-	-	-
6.7104	139000	0.0009	-	-	-
6.7346	139500	0.0009	-	-	-
6.7587	140000	0.0009	0.0004	0.7817	0.8555
6.7829	140500	0.0009	-	-	-
6.8070	141000	0.0009	-	-	-
6.8311	141500	0.0009	-	-	-
6.8553	142000	0.0009	0.0004	0.7812	0.8556
6.8794	142500	0.0009	-	-	-
6.9035	143000	0.0009	-	-	-
6.9277	143500	0.0009	-	-	-
6.9518	144000	0.0009	0.0004	0.7830	0.8559
6.9760	144500	0.0009	-	-	-
7.0001	145000	0.0009	-	-	-
7.0242	145500	0.0009	-	-	-
7.0484	146000	0.0009	0.0004	0.7809	0.8561
7.0725	146500	0.0009	-	-	-
7.0966	147000	0.0009	-	-	-
7.1208	147500	0.0009	-	-	-
7.1449	148000	0.0009	0.0004	0.7798	0.8560
7.1691	148500	0.0009	-	-	-
7.1932	149000	0.0009	-	-	-
7.2173	149500	0.0009	-	-	-
7.2415	150000	0.0009	0.0004	0.7815	0.8559
7.2656	150500	0.0009	-	-	-
7.2898	151000	0.0009	-	-	-
7.3139	151500	0.0009	-	-	-
7.3380	152000	0.0009	0.0004	0.7828	0.8562
7.3622	152500	0.0009	-	-	-
7.3863	153000	0.0009	-	-	-
7.4104	153500	0.0009	-	-	-
7.4346	154000	0.0009	0.0004	0.7837	0.8565
7.4587	154500	0.0009	-	-	-
7.4829	155000	0.0009	-	-	-
7.5070	155500	0.0009	-	-	-
7.5311	156000	0.0009	0.0004	0.7819	0.8565
7.5553	156500	0.0009	-	-	-
7.5794	157000	0.0009	-	-	-
7.6036	157500	0.0009	-	-	-
7.6277	158000	0.0009	0.0004	0.7818	0.8557
7.6518	158500	0.0009	-	-	-
7.6760	159000	0.0009	-	-	-
7.7001	159500	0.0009	-	-	-
7.7242	160000	0.0009	0.0004	0.7811	0.8557
7.7484	160500	0.0009	-	-	-
7.7725	161000	0.0009	-	-	-
7.7967	161500	0.0009	-	-	-
7.8208	162000	0.0009	0.0004	0.7821	0.8566
7.8449	162500	0.0009	-	-	-
7.8691	163000	0.0009	-	-	-
7.8932	163500	0.0009	-	-	-
7.9174	164000	0.0009	0.0004	0.7815	0.8564
7.9415	164500	0.0009	-	-	-
7.9656	165000	0.0009	-	-	-
7.9898	165500	0.0009	-	-	-

Framework Versions

Python: 3.10.16
Sentence Transformers: 3.3.1
Transformers: 4.51.3
PyTorch: 2.5.1+cu124
Accelerate: 1.2.1
Datasets: 3.2.0
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",
}

MSELoss

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