This is a sentence-transformers model finetuned from unsloth/bge-m3 on the augmented-olive-product-phonetic-sentence-wo-negative dataset. It maps sentences & paragraphs to a 1024-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: unsloth/bge-m3
Maximum Sequence Length: 8192 tokens
Output Dimensionality: 1024 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- augmented-olive-product-phonetic-sentence-wo-negative

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': 8192, 'do_lower_case': False, 'architecture': 'PeftModelForFeatureExtraction'})
  (1): Pooling({'word_embedding_dimension': 1024, '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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("dkqjrm/bge-m3-embedding-augmented-olive-phonetic-sentence-wo-negative-lora")
# Run inference
sentences = [
    '[S~2XL SIZE] 시크릿데이 순한코튼 슬림핏 입는 오버나이트 4종 택1',
    'オーバーナイト 2XL はどこ',
    '야돔 가격 좀',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000,  0.5315,  0.0021],
#         [ 0.5315,  1.0000, -0.0381],
#         [ 0.0021, -0.0381,  1.0000]])

Training Details

Training Dataset

augmented-olive-product-phonetic-sentence-wo-negative

Dataset: augmented-olive-product-phonetic-sentence-wo-negative at 7ff3a49
Size: 4,379,592 training samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 6 tokens
mean: 22.4 tokens
max: 59 tokens
min: 3 tokens
mean: 9.44 tokens
max: 40 tokens

Samples:

anchor	positive
`네이처메이드 슈퍼 RTG 오메가-3 900mg 40캡슐 (20일분)`	`自然制造欧米伽多少钱`
`[푸마] 여성 실키 터치 레이온 브라탑+팬티 블랙`	`레이온 브라탑 팬티`
`순수식품 루테인&오메가3 9종복합기능성 30캡슐 (1개월분)`	`price of 30 capsules lutein`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false 5}

Evaluation Dataset

augmented-olive-product-phonetic-sentence-wo-negative

Dataset: augmented-olive-product-phonetic-sentence-wo-negative at 7ff3a49
Size: 230,504 evaluation samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 7 tokens
mean: 22.16 tokens
max: 52 tokens
min: 3 tokens
mean: 9.23 tokens
max: 35 tokens

Samples:

anchor	positive
`암앤해머 컴플리트케어치약 150g`	`アームアンドハンマーコンプリートケアはどこ`
`[정신과의사조향/평정심] 이유즈 룸앤패브릭 스프레이 100ml 어웨어니스`	`이즈 룸앤패브릭 가격 좀`
`볼제너뮬러 유기농 크런치 400g 아몬드헤이즐넛`	`クランチシリアルいくら`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false 5}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
gradient_accumulation_steps: 16
learning_rate: 3e-05
num_train_epochs: 2
lr_scheduler_type: cosine
warmup_ratio: 0.1
fp16: True
push_to_hub: 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: 16
per_device_eval_batch_size: 8
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 16
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 3e-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: 2
max_steps: -1
lr_scheduler_type: cosine
lr_scheduler_kwargs: None
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
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}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
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: True
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
hub_revision: None
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: no
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: True
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss
0.0029	50	1.1928	-
0.0058	100	1.1371	-
0.0088	150	1.0197	-
0.0117	200	0.8587	-
0.0146	250	0.7119	-
0.0175	300	0.6037	-
0.0205	350	0.5407	-
0.0234	400	0.4777	-
0.0263	450	0.4353	-
0.0292	500	0.4172	-
0.0321	550	0.4154	-
0.0351	600	0.3924	-
0.0380	650	0.3805	-
0.0409	700	0.3584	-
0.0438	750	0.3495	-
0.0468	800	0.3423	-
0.0497	850	0.3306	-
0.0526	900	0.319	-
0.0555	950	0.3076	-
0.0585	1000	0.3121	-
0.0614	1050	0.2911	-
0.0643	1100	0.2952	-
0.0672	1150	0.2855	-
0.0701	1200	0.286	-
0.0731	1250	0.2731	-
0.0760	1300	0.265	-
0.0789	1350	0.2578	-
0.0818	1400	0.2539	-
0.0848	1450	0.245	-
0.0877	1500	0.2412	-
0.0906	1550	0.2369	-
0.0935	1600	0.2296	-
0.0964	1650	0.2313	-
0.0994	1700	0.2124	-
0.1023	1750	0.2212	-
0.1052	1800	0.2211	-
0.1081	1850	0.2244	-
0.1111	1900	0.2107	-
0.1140	1950	0.2013	-
0.1169	2000	0.2122	-
0.1198	2050	0.198	-
0.1228	2100	0.2041	-
0.1257	2150	0.2008	-
0.1286	2200	0.2013	-
0.1315	2250	0.1955	-
0.1344	2300	0.1931	-
0.1374	2350	0.1893	-
0.1403	2400	0.1852	-
0.1432	2450	0.1826	-
0.1461	2500	0.1773	-
0.1491	2550	0.1793	-
0.1520	2600	0.173	-
0.1549	2650	0.1781	-
0.1578	2700	0.1649	-
0.1607	2750	0.1716	-
0.1637	2800	0.1682	-
0.1666	2850	0.1686	-
0.1695	2900	0.1607	-
0.1724	2950	0.167	-
0.1754	3000	0.1584	0.0601
0.1783	3050	0.1582	-
0.1812	3100	0.1564	-
0.1841	3150	0.159	-
0.1870	3200	0.1541	-
0.1900	3250	0.1522	-
0.1929	3300	0.1566	-
0.1958	3350	0.1498	-
0.1987	3400	0.1573	-
0.2017	3450	0.1507	-
0.2046	3500	0.1482	-
0.2075	3550	0.1406	-
0.2104	3600	0.1512	-
0.2134	3650	0.143	-
0.2163	3700	0.1501	-
0.2192	3750	0.1446	-
0.2221	3800	0.1408	-
0.2250	3850	0.1448	-
0.2280	3900	0.1363	-
0.2309	3950	0.1336	-
0.2338	4000	0.1412	-
0.2367	4050	0.1382	-
0.2397	4100	0.1351	-
0.2426	4150	0.1359	-
0.2455	4200	0.1281	-
0.2484	4250	0.1246	-
0.2513	4300	0.1278	-
0.2543	4350	0.1231	-
0.2572	4400	0.1252	-
0.2601	4450	0.1221	-
0.2630	4500	0.1273	-
0.2660	4550	0.1193	-
0.2689	4600	0.1226	-
0.2718	4650	0.1255	-
0.2747	4700	0.1215	-
0.2777	4750	0.1141	-
0.2806	4800	0.1143	-
0.2835	4850	0.1217	-
0.2864	4900	0.1156	-
0.2893	4950	0.1162	-
0.2923	5000	0.1168	-
0.2952	5050	0.1108	-
0.2981	5100	0.1147	-
0.3010	5150	0.1082	-
0.3040	5200	0.1109	-
0.3069	5250	0.1113	-
0.3098	5300	0.1115	-
0.3127	5350	0.1117	-
0.3156	5400	0.1106	-
0.3186	5450	0.1144	-
0.3215	5500	0.0988	-
0.3244	5550	0.1104	-
0.3273	5600	0.1031	-
0.3303	5650	0.1076	-
0.3332	5700	0.1111	-
0.3361	5750	0.1005	-
0.3390	5800	0.1059	-
0.3419	5850	0.0994	-
0.3449	5900	0.1002	-
0.3478	5950	0.1012	-
0.3507	6000	0.0987	0.0351
0.3536	6050	0.1038	-
0.3566	6100	0.0985	-
0.3595	6150	0.096	-
0.3624	6200	0.0966	-
0.3653	6250	0.0942	-
0.3683	6300	0.0979	-
0.3712	6350	0.098	-
0.3741	6400	0.0959	-
0.3770	6450	0.0918	-
0.3799	6500	0.0941	-
0.3829	6550	0.0967	-
0.3858	6600	0.0948	-
0.3887	6650	0.0935	-
0.3916	6700	0.0939	-
0.3946	6750	0.093	-
0.3975	6800	0.0902	-
0.4004	6850	0.0944	-
0.4033	6900	0.0976	-
0.4062	6950	0.0951	-
0.4092	7000	0.0938	-
0.4121	7050	0.0849	-
0.4150	7100	0.0932	-
0.4179	7150	0.0914	-
0.4209	7200	0.0809	-
0.4238	7250	0.0933	-
0.4267	7300	0.093	-
0.4296	7350	0.0864	-
0.4326	7400	0.0877	-
0.4355	7450	0.0892	-
0.4384	7500	0.0838	-
0.4413	7550	0.0848	-
0.4442	7600	0.085	-
0.4472	7650	0.0874	-
0.4501	7700	0.0845	-
0.4530	7750	0.0824	-
0.4559	7800	0.0837	-
0.4589	7850	0.085	-
0.4618	7900	0.0862	-
0.4647	7950	0.0817	-
0.4676	8000	0.0836	-
0.4705	8050	0.0821	-
0.4735	8100	0.0848	-
0.4764	8150	0.0805	-
0.4793	8200	0.0795	-
0.4822	8250	0.0828	-
0.4852	8300	0.0817	-
0.4881	8350	0.084	-
0.4910	8400	0.0766	-
0.4939	8450	0.079	-
0.4968	8500	0.0779	-
0.4998	8550	0.0767	-
0.5027	8600	0.075	-
0.5056	8650	0.0809	-
0.5085	8700	0.0764	-
0.5115	8750	0.0787	-
0.5144	8800	0.0764	-
0.5173	8850	0.0824	-
0.5202	8900	0.0811	-
0.5232	8950	0.0756	-
0.5261	9000	0.0748	0.0263
0.5290	9050	0.0773	-
0.5319	9100	0.0762	-
0.5348	9150	0.0726	-
0.5378	9200	0.0747	-
0.5407	9250	0.0758	-
0.5436	9300	0.0799	-
0.5465	9350	0.0737	-
0.5495	9400	0.0753	-
0.5524	9450	0.0758	-
0.5553	9500	0.074	-
0.5582	9550	0.0798	-
0.5611	9600	0.0708	-
0.5641	9650	0.0729	-
0.5670	9700	0.0704	-
0.5699	9750	0.0742	-
0.5728	9800	0.0716	-
0.5758	9850	0.0731	-
0.5787	9900	0.0747	-
0.5816	9950	0.0705	-
0.5845	10000	0.0667	-
0.5875	10050	0.0719	-
0.5904	10100	0.0746	-
0.5933	10150	0.0671	-
0.5962	10200	0.0712	-
0.5991	10250	0.0714	-
0.6021	10300	0.0679	-
0.6050	10350	0.0722	-
0.6079	10400	0.0697	-
0.6108	10450	0.0676	-
0.6138	10500	0.0659	-
0.6167	10550	0.0726	-
0.6196	10600	0.0675	-
0.6225	10650	0.0693	-
0.6254	10700	0.0663	-
0.6284	10750	0.0642	-
0.6313	10800	0.0713	-
0.6342	10850	0.0699	-
0.6371	10900	0.0657	-
0.6401	10950	0.0669	-
0.6430	11000	0.0649	-
0.6459	11050	0.0693	-
0.6488	11100	0.0705	-
0.6517	11150	0.0662	-
0.6547	11200	0.0635	-
0.6576	11250	0.07	-
0.6605	11300	0.07	-
0.6634	11350	0.0683	-
0.6664	11400	0.0653	-
0.6693	11450	0.063	-
0.6722	11500	0.0627	-
0.6751	11550	0.0676	-
0.6781	11600	0.0664	-
0.6810	11650	0.0625	-
0.6839	11700	0.065	-
0.6868	11750	0.0647	-
0.6897	11800	0.063	-
0.6927	11850	0.066	-
0.6956	11900	0.0642	-
0.6985	11950	0.0604	-
0.7014	12000	0.0649	0.0218
0.7044	12050	0.0598	-
0.7073	12100	0.0646	-
0.7102	12150	0.0669	-
0.7131	12200	0.0634	-
0.7160	12250	0.0636	-
0.7190	12300	0.0634	-
0.7219	12350	0.0622	-
0.7248	12400	0.0674	-
0.7277	12450	0.0643	-
0.7307	12500	0.0659	-
0.7336	12550	0.062	-
0.7365	12600	0.0634	-
0.7394	12650	0.0627	-
0.7424	12700	0.0608	-
0.7453	12750	0.0578	-
0.7482	12800	0.0598	-
0.7511	12850	0.0627	-
0.7540	12900	0.0669	-
0.7570	12950	0.0621	-
0.7599	13000	0.0619	-
0.7628	13050	0.0596	-
0.7657	13100	0.0617	-
0.7687	13150	0.0634	-
0.7716	13200	0.0588	-
0.7745	13250	0.0597	-
0.7774	13300	0.061	-
0.7803	13350	0.0568	-
0.7833	13400	0.0626	-
0.7862	13450	0.0598	-
0.7891	13500	0.0601	-
0.7920	13550	0.0555	-
0.7950	13600	0.0593	-
0.7979	13650	0.0588	-
0.8008	13700	0.0652	-
0.8037	13750	0.0594	-
0.8066	13800	0.061	-
0.8096	13850	0.0594	-
0.8125	13900	0.0607	-
0.8154	13950	0.0603	-
0.8183	14000	0.0591	-
0.8213	14050	0.0598	-
0.8242	14100	0.0583	-
0.8271	14150	0.0583	-
0.8300	14200	0.0599	-
0.8330	14250	0.0572	-
0.8359	14300	0.0552	-
0.8388	14350	0.0552	-
0.8417	14400	0.058	-
0.8446	14450	0.0573	-
0.8476	14500	0.0609	-
0.8505	14550	0.059	-
0.8534	14600	0.0561	-
0.8563	14650	0.0556	-
0.8593	14700	0.0583	-
0.8622	14750	0.0548	-
0.8651	14800	0.0554	-
0.8680	14850	0.0619	-
0.8709	14900	0.0583	-
0.8739	14950	0.0539	-
0.8768	15000	0.0566	0.0189
0.8797	15050	0.0531	-
0.8826	15100	0.0561	-
0.8856	15150	0.0534	-
0.8885	15200	0.0501	-
0.8914	15250	0.0564	-
0.8943	15300	0.0545	-
0.8973	15350	0.0603	-
0.9002	15400	0.0547	-
0.9031	15450	0.058	-
0.9060	15500	0.0575	-
0.9089	15550	0.0601	-
0.9119	15600	0.0555	-
0.9148	15650	0.0542	-
0.9177	15700	0.0529	-
0.9206	15750	0.0541	-
0.9236	15800	0.0572	-
0.9265	15850	0.0531	-
0.9294	15900	0.057	-
0.9323	15950	0.0564	-
0.9352	16000	0.0506	-
0.9382	16050	0.0538	-
0.9411	16100	0.0537	-
0.9440	16150	0.0545	-
0.9469	16200	0.0558	-
0.9499	16250	0.0544	-
0.9528	16300	0.0528	-
0.9557	16350	0.0561	-
0.9586	16400	0.0552	-
0.9615	16450	0.0547	-
0.9645	16500	0.0549	-
0.9674	16550	0.0515	-
0.9703	16600	0.0512	-
0.9732	16650	0.0549	-
0.9762	16700	0.0537	-
0.9791	16750	0.0529	-
0.9820	16800	0.0531	-
0.9849	16850	0.0526	-
0.9879	16900	0.0524	-
0.9908	16950	0.0515	-
0.9937	17000	0.0519	-
0.9966	17050	0.0512	-
0.9995	17100	0.0489	-
1.0025	17150	0.0532	-
1.0054	17200	0.0497	-
1.0083	17250	0.0483	-
1.0112	17300	0.0505	-
1.0141	17350	0.0473	-
1.0171	17400	0.0513	-
1.0200	17450	0.049	-
1.0229	17500	0.0531	-
1.0258	17550	0.0472	-
1.0288	17600	0.0502	-
1.0317	17650	0.0517	-
1.0346	17700	0.0495	-
1.0375	17750	0.0474	-
1.0404	17800	0.0502	-
1.0434	17850	0.0498	-
1.0463	17900	0.0469	-
1.0492	17950	0.0503	-
1.0521	18000	0.0509	0.0166
1.0551	18050	0.0518	-
1.0580	18100	0.0525	-
1.0609	18150	0.0509	-
1.0638	18200	0.0515	-
1.0668	18250	0.0495	-
1.0697	18300	0.0505	-
1.0726	18350	0.0507	-
1.0755	18400	0.0533	-
1.0784	18450	0.0487	-
1.0814	18500	0.0492	-
1.0843	18550	0.05	-
1.0872	18600	0.0474	-
1.0901	18650	0.0514	-
1.0931	18700	0.0467	-
1.0960	18750	0.0488	-
1.0989	18800	0.0478	-
1.1018	18850	0.0485	-
1.1047	18900	0.0525	-
1.1077	18950	0.05	-
1.1106	19000	0.0472	-
1.1135	19050	0.048	-
1.1164	19100	0.0492	-
1.1194	19150	0.0475	-
1.1223	19200	0.0506	-
1.1252	19250	0.0474	-
1.1281	19300	0.0509	-
1.1311	19350	0.0474	-
1.1340	19400	0.0458	-
1.1369	19450	0.0459	-
1.1398	19500	0.0468	-
1.1427	19550	0.0471	-
1.1457	19600	0.0466	-
1.1486	19650	0.0461	-
1.1515	19700	0.0476	-
1.1544	19750	0.0464	-
1.1574	19800	0.0467	-
1.1603	19850	0.0462	-
1.1632	19900	0.0458	-
1.1661	19950	0.0465	-
1.1690	20000	0.0467	-
1.1720	20050	0.0509	-
1.1749	20100	0.0445	-
1.1778	20150	0.0461	-
1.1807	20200	0.0434	-
1.1837	20250	0.0469	-
1.1866	20300	0.0457	-
1.1895	20350	0.0477	-
1.1924	20400	0.0453	-
1.1953	20450	0.0445	-
1.1983	20500	0.0446	-
1.2012	20550	0.046	-
1.2041	20600	0.0475	-
1.2070	20650	0.0477	-
1.2100	20700	0.0481	-
1.2129	20750	0.0482	-
1.2158	20800	0.0451	-
1.2187	20850	0.0491	-
1.2217	20900	0.0456	-
1.2246	20950	0.0435	-
1.2275	21000	0.0449	0.0154
1.2304	21050	0.0489	-
1.2333	21100	0.0434	-
1.2363	21150	0.0449	-
1.2392	21200	0.0464	-
1.2421	21250	0.046	-
1.2450	21300	0.0465	-
1.2480	21350	0.0439	-
1.2509	21400	0.0462	-
1.2538	21450	0.0461	-
1.2567	21500	0.0456	-
1.2596	21550	0.0463	-
1.2626	21600	0.0431	-
1.2655	21650	0.0458	-
1.2684	21700	0.046	-
1.2713	21750	0.0438	-
1.2743	21800	0.0482	-
1.2772	21850	0.0438	-
1.2801	21900	0.0425	-
1.2830	21950	0.0442	-
1.2860	22000	0.0447	-
1.2889	22050	0.0434	-
1.2918	22100	0.0451	-
1.2947	22150	0.0423	-
1.2976	22200	0.0431	-
1.3006	22250	0.0426	-
1.3035	22300	0.0441	-
1.3064	22350	0.0462	-
1.3093	22400	0.044	-
1.3123	22450	0.0449	-
1.3152	22500	0.0461	-
1.3181	22550	0.0464	-
1.3210	22600	0.0424	-
1.3239	22650	0.0444	-
1.3269	22700	0.0412	-
1.3298	22750	0.0449	-
1.3327	22800	0.0465	-
1.3356	22850	0.0426	-
1.3386	22900	0.0429	-
1.3415	22950	0.0395	-
1.3444	23000	0.0451	-
1.3473	23050	0.042	-
1.3502	23100	0.0403	-
1.3532	23150	0.0451	-
1.3561	23200	0.0438	-
1.3590	23250	0.043	-
1.3619	23300	0.0418	-
1.3649	23350	0.0458	-
1.3678	23400	0.0441	-
1.3707	23450	0.0451	-
1.3736	23500	0.0445	-
1.3766	23550	0.0448	-
1.3795	23600	0.0402	-
1.3824	23650	0.0433	-
1.3853	23700	0.0437	-
1.3882	23750	0.046	-
1.3912	23800	0.0414	-
1.3941	23850	0.0409	-
1.3970	23900	0.0416	-
1.3999	23950	0.0438	-
1.4029	24000	0.0439	0.0145
1.4058	24050	0.0446	-
1.4087	24100	0.0459	-
1.4116	24150	0.0399	-
1.4145	24200	0.042	-
1.4175	24250	0.0421	-
1.4204	24300	0.0431	-
1.4233	24350	0.0424	-
1.4262	24400	0.0461	-
1.4292	24450	0.042	-
1.4321	24500	0.0452	-
1.4350	24550	0.0413	-
1.4379	24600	0.0433	-
1.4409	24650	0.0417	-
1.4438	24700	0.0418	-
1.4467	24750	0.0433	-
1.4496	24800	0.0431	-
1.4525	24850	0.0473	-
1.4555	24900	0.0393	-
1.4584	24950	0.0414	-
1.4613	25000	0.0432	-
1.4642	25050	0.045	-
1.4672	25100	0.0395	-
1.4701	25150	0.0416	-
1.4730	25200	0.0427	-
1.4759	25250	0.0447	-
1.4788	25300	0.0429	-
1.4818	25350	0.041	-
1.4847	25400	0.0428	-
1.4876	25450	0.0442	-
1.4905	25500	0.0423	-
1.4935	25550	0.0448	-
1.4964	25600	0.0425	-
1.4993	25650	0.0393	-
1.5022	25700	0.0408	-
1.5051	25750	0.0403	-
1.5081	25800	0.0411	-
1.5110	25850	0.0433	-
1.5139	25900	0.0419	-
1.5168	25950	0.0415	-
1.5198	26000	0.0417	-
1.5227	26050	0.0444	-
1.5256	26100	0.0419	-
1.5285	26150	0.0422	-
1.5315	26200	0.0419	-
1.5344	26250	0.0439	-
1.5373	26300	0.0416	-
1.5402	26350	0.0436	-
1.5431	26400	0.0463	-
1.5461	26450	0.0435	-
1.5490	26500	0.0422	-
1.5519	26550	0.0416	-
1.5548	26600	0.041	-
1.5578	26650	0.0409	-
1.5607	26700	0.0439	-
1.5636	26750	0.044	-
1.5665	26800	0.0437	-
1.5694	26850	0.043	-
1.5724	26900	0.0411	-
1.5753	26950	0.0421	-
1.5782	27000	0.0409	0.0140
1.5811	27050	0.0432	-
1.5841	27100	0.0421	-
1.5870	27150	0.0391	-
1.5899	27200	0.0418	-
1.5928	27250	0.0427	-
1.5958	27300	0.0428	-
1.5987	27350	0.0397	-
1.6016	27400	0.0437	-
1.6045	27450	0.0433	-
1.6074	27500	0.0394	-
1.6104	27550	0.0399	-
1.6133	27600	0.0427	-
1.6162	27650	0.0426	-
1.6191	27700	0.0438	-
1.6221	27750	0.0409	-
1.6250	27800	0.042	-
1.6279	27850	0.0406	-
1.6308	27900	0.0395	-
1.6337	27950	0.0406	-
1.6367	28000	0.043	-
1.6396	28050	0.0415	-
1.6425	28100	0.0405	-
1.6454	28150	0.0387	-
1.6484	28200	0.0408	-
1.6513	28250	0.0397	-
1.6542	28300	0.0443	-
1.6571	28350	0.0413	-
1.6600	28400	0.0423	-
1.6630	28450	0.0418	-
1.6659	28500	0.0424	-
1.6688	28550	0.0397	-
1.6717	28600	0.0415	-
1.6747	28650	0.0421	-
1.6776	28700	0.0409	-
1.6805	28750	0.0439	-
1.6834	28800	0.0426	-
1.6864	28850	0.0422	-
1.6893	28900	0.0413	-
1.6922	28950	0.0416	-
1.6951	29000	0.0374	-
1.6980	29050	0.0409	-
1.7010	29100	0.0412	-
1.7039	29150	0.041	-
1.7068	29200	0.0383	-
1.7097	29250	0.0423	-
1.7127	29300	0.0406	-
1.7156	29350	0.0422	-
1.7185	29400	0.038	-
1.7214	29450	0.04	-
1.7243	29500	0.0425	-
1.7273	29550	0.041	-
1.7302	29600	0.042	-
1.7331	29650	0.0432	-
1.7360	29700	0.0421	-
1.7390	29750	0.0394	-
1.7419	29800	0.0419	-
1.7448	29850	0.0411	-
1.7477	29900	0.0421	-
1.7507	29950	0.0392	-
1.7536	30000	0.0454	0.0138
1.7565	30050	0.041	-
1.7594	30100	0.0421	-
1.7623	30150	0.0417	-
1.7653	30200	0.0421	-
1.7682	30250	0.0375	-
1.7711	30300	0.0422	-
1.7740	30350	0.0412	-
1.7770	30400	0.0381	-
1.7799	30450	0.0441	-
1.7828	30500	0.0408	-
1.7857	30550	0.0396	-
1.7886	30600	0.04	-
1.7916	30650	0.0396	-
1.7945	30700	0.0427	-
1.7974	30750	0.0411	-
1.8003	30800	0.0432	-
1.8033	30850	0.0425	-
1.8062	30900	0.046	-
1.8091	30950	0.0404	-
1.8120	31000	0.0398	-
1.8149	31050	0.0428	-
1.8179	31100	0.0405	-
1.8208	31150	0.0404	-
1.8237	31200	0.0387	-
1.8266	31250	0.0412	-
1.8296	31300	0.0411	-
1.8325	31350	0.0426	-
1.8354	31400	0.0415	-
1.8383	31450	0.0407	-
1.8413	31500	0.0398	-
1.8442	31550	0.0439	-
1.8471	31600	0.0439	-
1.8500	31650	0.0397	-
1.8529	31700	0.0408	-
1.8559	31750	0.0393	-
1.8588	31800	0.043	-
1.8617	31850	0.0404	-
1.8646	31900	0.0414	-
1.8676	31950	0.0396	-
1.8705	32000	0.0427	-
1.8734	32050	0.0409	-
1.8763	32100	0.0395	-
1.8792	32150	0.0426	-
1.8822	32200	0.046	-
1.8851	32250	0.0388	-
1.8880	32300	0.0445	-
1.8909	32350	0.0408	-
1.8939	32400	0.0381	-
1.8968	32450	0.0421	-
1.8997	32500	0.0391	-
1.9026	32550	0.0412	-
1.9056	32600	0.041	-
1.9085	32650	0.0406	-
1.9114	32700	0.0409	-
1.9143	32750	0.0395	-
1.9172	32800	0.0388	-
1.9202	32850	0.0419	-
1.9231	32900	0.0418	-
1.9260	32950	0.0434	-
1.9289	33000	0.0397	0.0136
1.9319	33050	0.0389	-
1.9348	33100	0.0414	-
1.9377	33150	0.0388	-
1.9406	33200	0.0388	-
1.9435	33250	0.0415	-
1.9465	33300	0.0401	-
1.9494	33350	0.041	-
1.9523	33400	0.0412	-
1.9552	33450	0.0403	-
1.9582	33500	0.0425	-
1.9611	33550	0.0391	-
1.9640	33600	0.0399	-
1.9669	33650	0.0417	-
1.9698	33700	0.0401	-
1.9728	33750	0.0419	-
1.9757	33800	0.041	-
1.9786	33850	0.044	-
1.9815	33900	0.0386	-
1.9845	33950	0.0378	-
1.9874	34000	0.0388	-
1.9903	34050	0.0413	-
1.9932	34100	0.0408	-
1.9962	34150	0.0387	-
1.9991	34200	0.0386	-

Framework Versions

Python: 3.12.12
Sentence Transformers: 5.2.1
Transformers: 4.57.6
PyTorch: 2.10.0+cu128
Accelerate: 1.12.0
Datasets: 4.3.0
Tokenizers: 0.22.2

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}
}

	anchor	positive
type	string	string
details	min: 6 tokens mean: 22.4 tokens max: 59 tokens	min: 3 tokens mean: 9.44 tokens max: 40 tokens