metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:18240762
  - loss:MSELoss
base_model: sentence-transformers/all-MiniLM-L6-v2
widget:
  - source_sentence: Yeah, fire in the park, let's go!
    sentences:
      - 午前2時頃に音楽が止まり、それから熟睡。
      - 彼はニンジンが好きではないので、食べなかった。
      - 公園のライトアップ、ぜひ行こうね！
  - source_sentence: Population is around 5.7 million people.
    sentences:
      - 人口は約570万人です。
      - カンドンベの音楽はcuerdaと呼ばれるドラマーのグループによって演奏される。
      - 'シノプシス: 2116年—日本政府はシビルシステムの無人ドローンロボットを問題のある国に輸出し始め、システムは世界中に広がっています。'
  - source_sentence: >-
      With EMUI 5.0, the Huawei Mate 9 becomes more intelligent and efficient
      over time by understanding consumers’ behaviour patterns and ensures the
      highest priority applications are given preference subject to system
      resources.
    sentences:
      - 私も今はクルマを持っていません。
      - ガジュマルの樹を見に行きたいです。
      - >-
        EMUI5.0では、『HUAWEI Mate
        9』が消費者の行動パターンを理解し、時間をかけて知能と効率を上げ、優先順位の最も高いアプリをシステム消費源の対象に優先される事を保証します。
  - source_sentence: >-
      What are the differences between the environments and geographical
      positions of the East and the West?
    sentences:
      - 環境と地理的位置に関して、東洋と西洋の相違点は何であろうか。
      - >-
        その  ほか  に ， “心霊  手術  師 ” が  おり ， この  人  たち  は“ 心霊  手術 ” なる  もの
         を  行ない  ます。
      - Numpy を import できない。
  - source_sentence: Jesus Christ did surrender his life for the “sheep. ”
    sentences:
      - フィリポは読んでいる事柄が分かりますかと尋ねた。
      - イエス  ･  キリスト  は  ご自分  の  命  を「羊」の  ため  に  捨て  まし  た。
      - 彼はこの金を中央政府には渡そうとしない。
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: stsb multi mt en
          type: stsb_multi_mt-en
        metrics:
          - type: pearson_cosine
            value: 0.7901750255742193
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.793832704547488
            name: Spearman Cosine
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: JSTS
          type: JSTS
        metrics:
          - type: pearson_cosine
            value: 0.8562933057524594
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8081744506827298
            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': True, 'pooling_mode_mean_tokens': False, '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 = [
    'Jesus Christ did surrender his life for the “sheep. ”',
    'イエス \u200b ･ \u200b キリスト \u200b は \u200b ご自分 \u200b の \u200b 命 \u200b を「羊」の \u200b ため \u200b に \u200b 捨て \u200b まし \u200b た。',
    '彼はこの金を中央政府には渡そうとしない。',
]
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: stsb_multi_mt-en and JSTS
Evaluated with EmbeddingSimilarityEvaluator

Metric	stsb_multi_mt-en	JSTS
pearson_cosine	0.7902	0.8563
spearman_cosine	0.7938	0.8082

Training Details

Training Dataset

Unnamed Dataset

Size: 18,240,762 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: 15.99 tokens
max: 128 tokens

min: 4 tokens
mean: 21.59 tokens
max: 128 tokens

size: 384 elements

	english	non_english	label
type	string	string	list
details	min: 4 tokens mean: 15.99 tokens max: 128 tokens	min: 4 tokens mean: 21.59 tokens max: 128 tokens	size: 384 elements

Samples:

english	non_english	label
`Slow to Mars?`	`火星しばり?`	`[-0.1292940022648608, -0.1167307527589221, -0.008499974779641976, 0.04317784529767997, -0.06141806471633044, ...]`
`Sunset is nearly there.`	`サンクスはすぐそこだし。`	`[-0.1347740689698337, 0.053288680755846106, 0.014359346388162629, 0.0157641416547634, 0.0900218121125077, ...]`
`Why were these Christians put to death?`	`ハンガリーの新聞「バシュ･ナーペ」は次のように説明しています。「`	`[0.09746742956653999, -0.006846877375759926, -0.03973075126221857, 0.024986338940603363, -0.021140928354124164, ...]`

Loss: MSELoss

Evaluation Dataset

Unnamed Dataset

Size: 184,251 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: 16.16 tokens
max: 116 tokens

min: 4 tokens
mean: 21.65 tokens
max: 128 tokens

size: 384 elements

	english	non_english	label
type	string	string	list
details	min: 4 tokens mean: 16.16 tokens max: 116 tokens	min: 4 tokens mean: 21.65 tokens max: 128 tokens	size: 384 elements

Samples:

english	non_english	label
`Back from donating?`	`ドーナツ回？`	`[-0.14056862827741115, -0.09391276023432168, 0.011405737148041988, 0.012085375305688852, -0.056379213184557624, ...]`
`134)Textbooks were also in short supply.`	`3）荷物の引き渡しも短時間にテキパキとされていました。`	`[0.04401202896633807, 0.07403046630916377, 0.11568493170920714, 0.047522982370575784, 0.1009405093401555, ...]`
`The COG investigators started the trial by providing dosages of crizotinib to their patients that were lower than those used in adults with NSCLC.`	`COG試験責任医師らは、NSCLCの成人患者で使用されている投与量より少ない量のcrizotinibを小児患者に提供することで試験を開始した。`	`[0.21476626448171793, -0.04704800523318936, 0.061019190603563075, 0.027317017405848458, -0.03788587912458321, ...]`

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	stsb_multi_mt-en_spearman_cosine	JSTS_spearman_cosine
0.0281	500	0.0058	-	-	-
0.0561	1000	0.0051	-	-	-
0.0842	1500	0.0048	-	-	-
0.1123	2000	0.0046	0.0022	0.2515	0.2726
0.1403	2500	0.0044	-	-	-
0.1684	3000	0.0043	-	-	-
0.1965	3500	0.0041	-	-	-
0.2245	4000	0.004	0.0019	0.4462	0.4910
0.2526	4500	0.0039	-	-	-
0.2807	5000	0.0038	-	-	-
0.3088	5500	0.0037	-	-	-
0.3368	6000	0.0036	0.0017	0.5792	0.6327
0.3649	6500	0.0035	-	-	-
0.3930	7000	0.0034	-	-	-
0.4210	7500	0.0033	-	-	-
0.4491	8000	0.0032	0.0015	0.6501	0.7016
0.4772	8500	0.0032	-	-	-
0.5052	9000	0.0031	-	-	-
0.5333	9500	0.0031	-	-	-
0.5614	10000	0.0031	0.0015	0.6939	0.7373
0.5894	10500	0.003	-	-	-
0.6175	11000	0.003	-	-	-
0.6456	11500	0.003	-	-	-
0.6736	12000	0.0029	0.0014	0.7043	0.7573
0.7017	12500	0.0029	-	-	-
0.7298	13000	0.0029	-	-	-
0.7579	13500	0.0029	-	-	-
0.7859	14000	0.0029	0.0014	0.7221	0.7642
0.8140	14500	0.0028	-	-	-
0.8421	15000	0.0028	-	-	-
0.8701	15500	0.0028	-	-	-
0.8982	16000	0.0028	0.0013	0.7400	0.7763
0.9263	16500	0.0028	-	-	-
0.9543	17000	0.0028	-	-	-
0.9824	17500	0.0028	-	-	-
1.0104	18000	0.0027	0.0013	0.7459	0.7796
1.0385	18500	0.0027	-	-	-
1.0666	19000	0.0027	-	-	-
1.0946	19500	0.0027	-	-	-
1.1227	20000	0.0027	0.0013	0.7620	0.7853
1.1508	20500	0.0027	-	-	-
1.1789	21000	0.0027	-	-	-
1.2069	21500	0.0027	-	-	-
1.2350	22000	0.0027	0.0013	0.7669	0.7848
1.2631	22500	0.0027	-	-	-
1.2911	23000	0.0027	-	-	-
1.3192	23500	0.0026	-	-	-
1.3473	24000	0.0026	0.0013	0.7633	0.7866
1.3753	24500	0.0026	-	-	-
1.4034	25000	0.0026	-	-	-
1.4315	25500	0.0026	-	-	-
1.4595	26000	0.0026	0.0013	0.7751	0.7892
1.4876	26500	0.0026	-	-	-
1.5157	27000	0.0026	-	-	-
1.5437	27500	0.0026	-	-	-
1.5718	28000	0.0026	0.0012	0.7751	0.7951
1.5999	28500	0.0026	-	-	-
1.6280	29000	0.0026	-	-	-
1.6560	29500	0.0026	-	-	-
1.6841	30000	0.0026	0.0012	0.7765	0.7957
1.7122	30500	0.0026	-	-	-
1.7402	31000	0.0026	-	-	-
1.7683	31500	0.0026	-	-	-
1.7964	32000	0.0026	0.0012	0.7805	0.7957
1.8244	32500	0.0026	-	-	-
1.8525	33000	0.0026	-	-	-
1.8806	33500	0.0026	-	-	-
1.9086	34000	0.0026	0.0012	0.7797	0.7958
1.9367	34500	0.0026	-	-	-
1.9648	35000	0.0026	-	-	-
1.9928	35500	0.0026	-	-	-
2.0209	36000	0.0025	0.0012	0.7792	0.7943
2.0490	36500	0.0025	-	-	-
2.0770	37000	0.0025	-	-	-
2.1051	37500	0.0025	-	-	-
2.1332	38000	0.0025	0.0012	0.7831	0.7943
2.1612	38500	0.0025	-	-	-
2.1893	39000	0.0025	-	-	-
2.2174	39500	0.0025	-	-	-
2.2454	40000	0.0025	0.0012	0.7834	0.7973
2.2735	40500	0.0025	-	-	-
2.3016	41000	0.0025	-	-	-
2.3296	41500	0.0025	-	-	-
2.3577	42000	0.0025	0.0012	0.7860	0.7988
2.3858	42500	0.0025	-	-	-
2.4138	43000	0.0025	-	-	-
2.4419	43500	0.0025	-	-	-
2.4700	44000	0.0025	0.0012	0.7867	0.8006
2.4980	44500	0.0025	-	-	-
2.5261	45000	0.0025	-	-	-
2.5542	45500	0.0025	-	-	-
2.5823	46000	0.0025	0.0012	0.7870	0.8009
2.6103	46500	0.0025	-	-	-
2.6384	47000	0.0025	-	-	-
2.6665	47500	0.0025	-	-	-
2.6945	48000	0.0025	0.0012	0.7852	0.8019
2.7226	48500	0.0025	-	-	-
2.7507	49000	0.0025	-	-	-
2.7787	49500	0.0025	-	-	-
2.8068	50000	0.0025	0.0012	0.7863	0.8018
2.8349	50500	0.0025	-	-	-
2.8629	51000	0.0025	-	-	-
2.8910	51500	0.0025	-	-	-
2.9191	52000	0.0025	0.0012	0.7874	0.8000
2.9471	52500	0.0025	-	-	-
2.9752	53000	0.0025	-	-	-
3.0033	53500	0.0025	-	-	-
3.0313	54000	0.0025	0.0012	0.7875	0.8007
3.0594	54500	0.0025	-	-	-
3.0875	55000	0.0025	-	-	-
3.1155	55500	0.0025	-	-	-
3.1436	56000	0.0025	0.0012	0.7899	0.8021
3.1717	56500	0.0025	-	-	-
3.1997	57000	0.0025	-	-	-
3.2278	57500	0.0025	-	-	-
3.2559	58000	0.0025	0.0012	0.7914	0.8014
3.2839	58500	0.0025	-	-	-
3.3120	59000	0.0025	-	-	-
3.3401	59500	0.0025	-	-	-
3.3681	60000	0.0025	0.0012	0.7860	0.8029
3.3962	60500	0.0025	-	-	-
3.4243	61000	0.0025	-	-	-
3.4524	61500	0.0025	-	-	-
3.4804	62000	0.0025	0.0012	0.7886	0.8023
3.5085	62500	0.0025	-	-	-
3.5366	63000	0.0025	-	-	-
3.5646	63500	0.0025	-	-	-
3.5927	64000	0.0025	0.0012	0.7891	0.8045
3.6208	64500	0.0025	-	-	-
3.6488	65000	0.0025	-	-	-
3.6769	65500	0.0025	-	-	-
3.7050	66000	0.0025	0.0012	0.7892	0.8042
3.7330	66500	0.0025	-	-	-
3.7611	67000	0.0025	-	-	-
3.7892	67500	0.0025	-	-	-
3.8172	68000	0.0025	0.0012	0.7881	0.8042
3.8453	68500	0.0025	-	-	-
3.8734	69000	0.0025	-	-	-
3.9015	69500	0.0025	-	-	-
3.9295	70000	0.0025	0.0012	0.7905	0.8038
3.9576	70500	0.0025	-	-	-
3.9857	71000	0.0025	-	-	-
4.0137	71500	0.0025	-	-	-
4.0418	72000	0.0025	0.0012	0.7900	0.8052
4.0698	72500	0.0025	-	-	-
4.0979	73000	0.0025	-	-	-
4.1260	73500	0.0025	-	-	-
4.1540	74000	0.0025	0.0012	0.7904	0.8058
4.1821	74500	0.0025	-	-	-
4.2102	75000	0.0025	-	-	-
4.2382	75500	0.0025	-	-	-
4.2663	76000	0.0025	0.0012	0.7873	0.8049
4.2944	76500	0.0025	-	-	-
4.3225	77000	0.0025	-	-	-
4.3505	77500	0.0025	-	-	-
4.3786	78000	0.0025	0.0012	0.7908	0.8064
4.4067	78500	0.0025	-	-	-
4.4347	79000	0.0025	-	-	-
4.4628	79500	0.0025	-	-	-
4.4909	80000	0.0025	0.0012	0.7894	0.8050
4.5189	80500	0.0025	-	-	-
4.5470	81000	0.0025	-	-	-
4.5751	81500	0.0025	-	-	-
4.6031	82000	0.0025	0.0012	0.7917	0.8075
4.6312	82500	0.0025	-	-	-
4.6593	83000	0.0025	-	-	-
4.6873	83500	0.0025	-	-	-
4.7154	84000	0.0025	0.0012	0.7914	0.8059
4.7435	84500	0.0025	-	-	-
4.7715	85000	0.0025	-	-	-
4.7996	85500	0.0025	-	-	-
4.8277	86000	0.0025	0.0012	0.7895	0.8056
4.8558	86500	0.0025	-	-	-
4.8838	87000	0.0025	-	-	-
4.9119	87500	0.0025	-	-	-
4.9400	88000	0.0025	0.0012	0.7904	0.8059
4.9680	88500	0.0025	-	-	-
4.9961	89000	0.0025	-	-	-
5.0241	89500	0.0025	-	-	-
5.0522	90000	0.0025	0.0012	0.7907	0.8055
5.0803	90500	0.0025	-	-	-
5.1083	91000	0.0025	-	-	-
5.1364	91500	0.0025	-	-	-
5.1645	92000	0.0025	0.0012	0.7912	0.8056
5.1926	92500	0.0025	-	-	-
5.2206	93000	0.0025	-	-	-
5.2487	93500	0.0024	-	-	-
5.2768	94000	0.0025	0.0012	0.7913	0.8045
5.3048	94500	0.0025	-	-	-
5.3329	95000	0.0024	-	-	-
5.3610	95500	0.0024	-	-	-
5.3890	96000	0.0024	0.0012	0.7922	0.8056
5.4171	96500	0.0024	-	-	-
5.4452	97000	0.0024	-	-	-
5.4732	97500	0.0024	-	-	-
5.5013	98000	0.0024	0.0012	0.7909	0.8056
5.5294	98500	0.0024	-	-	-
5.5574	99000	0.0024	-	-	-
5.5855	99500	0.0024	-	-	-
5.6136	100000	0.0024	0.0012	0.7912	0.8075
5.6416	100500	0.0024	-	-	-
5.6697	101000	0.0024	-	-	-
5.6978	101500	0.0024	-	-	-
5.7259	102000	0.0024	0.0012	0.7921	0.8066
5.7539	102500	0.0024	-	-	-
5.7820	103000	0.0024	-	-	-
5.8101	103500	0.0024	-	-	-
5.8381	104000	0.0024	0.0012	0.7923	0.8068
5.8662	104500	0.0024	-	-	-
5.8943	105000	0.0024	-	-	-
5.9223	105500	0.0024	-	-	-
5.9504	106000	0.0024	0.0012	0.7941	0.8070
5.9785	106500	0.0024	-	-	-
6.0065	107000	0.0024	-	-	-
6.0346	107500	0.0024	-	-	-
6.0626	108000	0.0024	0.0012	0.7922	0.8078
6.0907	108500	0.0024	-	-	-
6.1188	109000	0.0024	-	-	-
6.1469	109500	0.0024	-	-	-
6.1749	110000	0.0024	0.0012	0.7922	0.8064
6.2030	110500	0.0024	-	-	-
6.2311	111000	0.0024	-	-	-
6.2591	111500	0.0024	-	-	-
6.2872	112000	0.0024	0.0012	0.7915	0.8069
6.3153	112500	0.0024	-	-	-
6.3433	113000	0.0024	-	-	-
6.3714	113500	0.0024	-	-	-
6.3995	114000	0.0024	0.0012	0.7921	0.8079
6.4275	114500	0.0024	-	-	-
6.4556	115000	0.0024	-	-	-
6.4837	115500	0.0024	-	-	-
6.5117	116000	0.0024	0.0012	0.7915	0.8071
6.5398	116500	0.0024	-	-	-
6.5679	117000	0.0024	-	-	-
6.5960	117500	0.0024	-	-	-
6.6240	118000	0.0024	0.0012	0.7943	0.8081
6.6521	118500	0.0024	-	-	-
6.6802	119000	0.0024	-	-	-
6.7082	119500	0.0024	-	-	-
6.7363	120000	0.0024	0.0012	0.7946	0.8079
6.7644	120500	0.0024	-	-	-
6.7924	121000	0.0024	-	-	-
6.8205	121500	0.0024	-	-	-
6.8486	122000	0.0024	0.0012	0.7919	0.8077
6.8766	122500	0.0024	-	-	-
6.9047	123000	0.0024	-	-	-
6.9328	123500	0.0024	-	-	-
6.9608	124000	0.0024	0.0012	0.7950	0.8087
6.9889	124500	0.0024	-	-	-
7.0170	125000	0.0024	-	-	-
7.0450	125500	0.0024	-	-	-
7.0731	126000	0.0024	0.0012	0.7927	0.8081
7.1012	126500	0.0024	-	-	-
7.1292	127000	0.0024	-	-	-
7.1573	127500	0.0024	-	-	-
7.1854	128000	0.0024	0.0012	0.7945	0.8082
7.2134	128500	0.0024	-	-	-
7.2415	129000	0.0024	-	-	-
7.2696	129500	0.0024	-	-	-
7.2976	130000	0.0024	0.0012	0.7927	0.8074
7.3257	130500	0.0024	-	-	-
7.3538	131000	0.0024	-	-	-
7.3818	131500	0.0024	-	-	-
7.4099	132000	0.0024	0.0012	0.7924	0.8077
7.4380	132500	0.0024	-	-	-
7.4661	133000	0.0024	-	-	-
7.4941	133500	0.0024	-	-	-
7.5222	134000	0.0024	0.0012	0.7929	0.8082
7.5503	134500	0.0024	-	-	-
7.5783	135000	0.0024	-	-	-
7.6064	135500	0.0024	-	-	-
7.6345	136000	0.0024	0.0012	0.7937	0.8080
7.6625	136500	0.0024	-	-	-
7.6906	137000	0.0024	-	-	-
7.7187	137500	0.0024	-	-	-
7.7467	138000	0.0024	0.0012	0.7941	0.8083
7.7748	138500	0.0024	-	-	-
7.8029	139000	0.0024	-	-	-
7.8309	139500	0.0024	-	-	-
7.8590	140000	0.0024	0.0012	0.7943	0.8082
7.8871	140500	0.0024	-	-	-
7.9151	141000	0.0024	-	-	-
7.9432	141500	0.0024	-	-	-
7.9713	142000	0.0024	0.0012	0.7938	0.8082
7.9994	142500	0.0024	-	-	-

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",
}