SentenceTransformer

This is a sentence-transformers model trained on the generator dataset. It maps sentences & paragraphs to a 4096-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
Maximum Sequence Length: 32768 tokens
Output Dimensionality: 4096 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- generator

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': 32768, 'do_lower_case': False, 'architecture': 'Qwen3Model'})
  (1): Pooling({'word_embedding_dimension': 2048, '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': True, '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("reasonwang/embedding-qwen3-1.7b-embedding_ctxt_norm-ac_unicode_shuf")
# Run inference
sentences = [
    'The weather is lovely today.',
    "It's so sunny outside!",
    'He drove to the stadium.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 4096]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.8363, 0.8542],
#         [0.8363, 1.0000, 0.7354],
#         [0.8542, 0.7354, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: validation_retrieval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.57
cosine_accuracy@3	0.73
cosine_accuracy@5	0.82
cosine_accuracy@10	0.87
cosine_precision@1	0.57
cosine_precision@3	0.4267
cosine_precision@5	0.388
cosine_precision@10	0.287
cosine_recall@1	0.1138
cosine_recall@3	0.2154
cosine_recall@5	0.279
cosine_recall@10	0.3847
cosine_ndcg@10	0.4588
cosine_ndcg@100	0.5435
cosine_mrr@10	0.6694
cosine_mrr@100	0.6763
cosine_map@100	0.3468

Training Details

Training Dataset

generator

Dataset: generator
Columns: sentence1 and sentence2

Loss: CachedMultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim",
    "mini_batch_size": 4,
    "gather_across_devices": false
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 256
learning_rate: 2e-05
max_steps: 100000
log_level: info
bf16: True
dataloader_num_workers: 1
accelerator_config: {'split_batches': False, 'dispatch_batches': False, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}

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: 8
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: 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: 3.0
max_steps: 100000
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
warmup_steps: 0
log_level: info
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: 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: True
dataloader_num_workers: 1
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': False, '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: False
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: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	validation_retrieval_cosine_ndcg@100
1e-05	1	5.3176	-
0.0001	10	3.7142	-
0.0002	20	2.7244	-
0.0003	30	2.4237	-
0.0004	40	2.2885	-
0.0005	50	2.2262	-
0.0006	60	2.1749	-
0.0007	70	2.1354	-
0.0008	80	2.1203	-
0.0009	90	2.1046	-
0.001	100	2.0212	0.4833
0.0011	110	2.0634	-
0.0012	120	2.0314	-
0.0013	130	2.0169	-
0.0014	140	2.0084	-
0.0015	150	1.9816	-
0.0016	160	1.9899	-
0.0017	170	1.9719	-
0.0018	180	1.9679	-
0.0019	190	1.9551	-
0.002	200	1.9742	0.4850
0.0021	210	1.941	-
0.0022	220	1.9453	-
0.0023	230	1.9353	-
0.0024	240	1.9497	-
0.0025	250	1.9494	-
0.0026	260	1.913	-
0.0027	270	1.8828	-
0.0028	280	1.9095	-
0.0029	290	1.8569	-
0.003	300	1.8816	0.4975
0.0031	310	1.8744	-
0.0032	320	1.8659	-
0.0033	330	1.8839	-
0.0034	340	1.8657	-
0.0035	350	1.8466	-
0.0036	360	1.8522	-
0.0037	370	1.8818	-
0.0038	380	1.8514	-
0.0039	390	1.8587	-
0.004	400	1.8437	0.5069
0.0041	410	1.8355	-
0.0042	420	1.8383	-
0.0043	430	1.8217	-
0.0044	440	1.8255	-
0.0045	450	1.7957	-
0.0046	460	1.8525	-
0.0047	470	1.8276	-
0.0048	480	1.8013	-
0.0049	490	1.781	-
0.005	500	1.7994	0.5042
0.0051	510	1.8029	-
0.0052	520	1.8223	-
0.0053	530	1.8072	-
0.0054	540	1.8269	-
0.0055	550	1.7999	-
0.0056	560	1.7895	-
0.0057	570	1.7911	-
0.0058	580	1.7864	-
0.0059	590	1.7952	-
0.006	600	1.7698	0.5078
0.0061	610	1.7969	-
0.0062	620	1.7678	-
0.0063	630	1.7823	-
0.0064	640	1.7613	-
0.0065	650	1.7638	-
0.0066	660	1.7755	-
0.0067	670	1.7283	-
0.0068	680	1.7642	-
0.0069	690	1.7748	-
0.007	700	1.763	0.5038
0.0071	710	1.7699	-
0.0072	720	1.7459	-
0.0073	730	1.7652	-
0.0074	740	1.7453	-
0.0075	750	1.77	-
0.0076	760	1.7641	-
0.0077	770	1.7652	-
0.0078	780	1.7633	-
0.0079	790	1.7224	-
0.008	800	1.726	0.5130
0.0081	810	1.7314	-
0.0082	820	1.75	-
0.0083	830	1.7203	-
0.0084	840	1.7266	-
0.0085	850	1.7358	-
0.0086	860	1.7346	-
0.0087	870	1.7602	-
0.0088	880	1.7229	-
0.0089	890	1.7382	-
0.009	900	1.7359	0.5093
0.0091	910	1.7299	-
0.0092	920	1.7252	-
0.0093	930	1.722	-
0.0094	940	1.7261	-
0.0095	950	1.7174	-
0.0096	960	1.7301	-
0.0097	970	1.7192	-
0.0098	980	1.7077	-
0.0099	990	1.7057	-
0.01	1000	1.7373	0.5107
0.0101	1010	1.718	-
0.0102	1020	1.7149	-
0.0103	1030	1.7054	-
0.0104	1040	1.7199	-
0.0105	1050	1.6925	-
0.0106	1060	1.7336	-
0.0107	1070	1.6887	-
0.0108	1080	1.707	-
0.0109	1090	1.6936	-
0.011	1100	1.6904	0.5120
0.0111	1110	1.7043	-
0.0112	1120	1.6975	-
0.0113	1130	1.7226	-
0.0114	1140	1.7069	-
0.0115	1150	1.696	-
0.0116	1160	1.6809	-
0.0117	1170	1.6954	-
0.0118	1180	1.6882	-
0.0119	1190	1.6952	-
0.012	1200	1.7232	0.5269
0.0121	1210	1.705	-
0.0122	1220	1.6382	-
0.0123	1230	1.6644	-
0.0124	1240	1.6798	-
0.0125	1250	1.7063	-
0.0126	1260	1.6858	-
0.0127	1270	1.689	-
0.0128	1280	1.6831	-
0.0129	1290	1.685	-
0.013	1300	1.7158	0.5148
0.0131	1310	1.6856	-
0.0132	1320	1.6545	-
0.0133	1330	1.6744	-
0.0134	1340	1.6673	-
0.0135	1350	1.664	-
0.0136	1360	1.6647	-
0.0137	1370	1.6636	-
0.0138	1380	1.6574	-
0.0139	1390	1.6949	-
0.014	1400	1.6896	0.5209
0.0141	1410	1.6497	-
0.0142	1420	1.6759	-
0.0143	1430	1.6894	-
0.0144	1440	1.656	-
0.0145	1450	1.644	-
0.0146	1460	1.6475	-
0.0147	1470	1.6525	-
0.0148	1480	1.6574	-
0.0149	1490	1.6741	-
0.015	1500	1.6541	0.5238
0.0151	1510	1.642	-
0.0152	1520	1.6588	-
0.0153	1530	1.6566	-
0.0154	1540	1.6603	-
0.0155	1550	1.6831	-
0.0156	1560	1.6464	-
0.0157	1570	1.6417	-
0.0158	1580	1.656	-
0.0159	1590	1.621	-
0.016	1600	1.6475	0.5211
0.0161	1610	1.6726	-
0.0162	1620	1.6196	-
0.0163	1630	1.6479	-
0.0164	1640	1.655	-
0.0165	1650	1.6515	-
0.0166	1660	1.635	-
0.0167	1670	1.6374	-
0.0168	1680	1.6586	-
0.0169	1690	1.6379	-
0.017	1700	1.6453	0.5158
0.0171	1710	1.6462	-
0.0172	1720	1.6465	-
0.0173	1730	1.6228	-
0.0174	1740	1.6458	-
0.0175	1750	1.6342	-
0.0176	1760	1.6255	-
0.0177	1770	1.6214	-
0.0178	1780	1.6426	-
0.0179	1790	1.6281	-
0.018	1800	1.6188	0.5233
0.0181	1810	1.6502	-
0.0182	1820	1.6285	-
0.0183	1830	1.6125	-
0.0184	1840	1.648	-
0.0185	1850	1.6458	-
0.0186	1860	1.6157	-
0.0187	1870	1.6634	-
0.0188	1880	1.6428	-
0.0189	1890	1.6304	-
0.019	1900	1.6278	0.5225
0.0191	1910	1.6177	-
0.0192	1920	1.6137	-
0.0193	1930	1.6231	-
0.0194	1940	1.6361	-
0.0195	1950	1.6307	-
0.0196	1960	1.6429	-
0.0197	1970	1.6463	-
0.0198	1980	1.6012	-
0.0199	1990	1.6212	-
0.02	2000	1.6157	0.5208
0.0201	2010	1.6181	-
0.0202	2020	1.6366	-
0.0203	2030	1.6401	-
0.0204	2040	1.6149	-
0.0205	2050	1.6142	-
0.0206	2060	1.6188	-
0.0207	2070	1.6132	-
0.0208	2080	1.6156	-
0.0209	2090	1.6094	-
0.021	2100	1.5963	0.5214
0.0211	2110	1.6325	-
0.0212	2120	1.6087	-
0.0213	2130	1.6155	-
0.0214	2140	1.5913	-
0.0215	2150	1.5951	-
0.0216	2160	1.602	-
0.0217	2170	1.598	-
0.0218	2180	1.6068	-
0.0219	2190	1.6018	-
0.022	2200	1.6178	0.5304
0.0221	2210	1.6123	-
0.0222	2220	1.6054	-
0.0223	2230	1.5984	-
0.0224	2240	1.5841	-
0.0225	2250	1.6093	-
0.0226	2260	1.6094	-
0.0227	2270	1.6175	-
0.0228	2280	1.6213	-
0.0229	2290	1.5803	-
0.023	2300	1.6042	0.5240
0.0231	2310	1.5669	-
0.0232	2320	1.6003	-
0.0233	2330	1.6133	-
0.0234	2340	1.611	-
0.0235	2350	1.5939	-
0.0236	2360	1.5942	-
0.0237	2370	1.5673	-
0.0238	2380	1.5723	-
0.0239	2390	1.5951	-
0.024	2400	1.5988	0.5306
0.0241	2410	1.5795	-
0.0242	2420	1.5766	-
0.0243	2430	1.5971	-
0.0244	2440	1.5764	-
0.0245	2450	1.5912	-
0.0246	2460	1.6069	-
0.0247	2470	1.5937	-
0.0248	2480	1.5698	-
0.0249	2490	1.5945	-
0.025	2500	1.6127	0.5317
0.0251	2510	1.5962	-
0.0252	2520	1.6053	-
0.0253	2530	1.5987	-
0.0254	2540	1.6135	-
0.0255	2550	1.5998	-
0.0256	2560	1.573	-
0.0257	2570	1.5772	-
0.0258	2580	1.598	-
0.0259	2590	1.5588	-
0.026	2600	1.5777	0.5221
0.0261	2610	1.5892	-
0.0262	2620	1.5648	-
0.0263	2630	1.5936	-
0.0264	2640	1.5931	-
0.0265	2650	1.5911	-
0.0266	2660	1.5731	-
0.0267	2670	1.5677	-
0.0268	2680	1.5933	-
0.0269	2690	1.596	-
0.027	2700	1.5879	0.5220
0.0271	2710	1.5605	-
0.0272	2720	1.5674	-
0.0273	2730	1.5871	-
0.0274	2740	1.5845	-
0.0275	2750	1.587	-
0.0276	2760	1.5997	-
0.0277	2770	1.5697	-
0.0278	2780	1.5773	-
0.0279	2790	1.5954	-
0.028	2800	1.5756	0.5270
0.0281	2810	1.5834	-
0.0282	2820	1.571	-
0.0283	2830	1.558	-
0.0284	2840	1.5785	-
0.0285	2850	1.5687	-
0.0286	2860	1.5795	-
0.0287	2870	1.5998	-
0.0288	2880	1.555	-
0.0289	2890	1.5643	-
0.029	2900	1.5679	0.5301
0.0291	2910	1.5638	-
0.0292	2920	1.5663	-
0.0293	2930	1.5759	-
0.0294	2940	1.5935	-
0.0295	2950	1.5652	-
0.0296	2960	1.5631	-
0.0297	2970	1.5646	-
0.0298	2980	1.5754	-
0.0299	2990	1.5577	-
0.03	3000	1.565	0.5296
0.0301	3010	1.5746	-
0.0302	3020	1.5526	-
0.0303	3030	1.5788	-
0.0304	3040	1.572	-
0.0305	3050	1.563	-
0.0306	3060	1.5704	-
0.0307	3070	1.5409	-
0.0308	3080	1.5709	-
0.0309	3090	1.5405	-
0.031	3100	1.5872	0.5253
0.0311	3110	1.5467	-
0.0312	3120	1.5502	-
0.0313	3130	1.549	-
0.0314	3140	1.5878	-
0.0315	3150	1.5582	-
0.0316	3160	1.5685	-
0.0317	3170	1.5718	-
0.0318	3180	1.5645	-
0.0319	3190	1.5678	-
0.032	3200	1.5752	0.5219
0.0321	3210	1.5528	-
0.0322	3220	1.5565	-
0.0323	3230	1.553	-
0.0324	3240	1.5712	-
0.0325	3250	1.5704	-
0.0326	3260	1.5453	-
0.0327	3270	1.5432	-
0.0328	3280	1.5529	-
0.0329	3290	1.58	-
0.033	3300	1.5593	0.5296
0.0331	3310	1.5757	-
0.0332	3320	1.5646	-
0.0333	3330	1.5502	-
0.0334	3340	1.5581	-
0.0335	3350	1.5708	-
0.0336	3360	1.5768	-
0.0337	3370	1.539	-
0.0338	3380	1.5683	-
0.0339	3390	1.5601	-
0.034	3400	1.5572	0.5293
0.0341	3410	1.5497	-
0.0342	3420	1.5647	-
0.0343	3430	1.552	-
0.0344	3440	1.5444	-
0.0345	3450	1.5396	-
0.0346	3460	1.553	-
0.0347	3470	1.5506	-
0.0348	3480	1.5524	-
0.0349	3490	1.5644	-
0.035	3500	1.529	0.5317
0.0351	3510	1.541	-
0.0352	3520	1.5446	-
0.0353	3530	1.5612	-
0.0354	3540	1.5459	-
0.0355	3550	1.5433	-
0.0356	3560	1.5437	-
0.0357	3570	1.5824	-
0.0358	3580	1.5373	-
0.0359	3590	1.5503	-
0.036	3600	1.5351	0.5390
0.0361	3610	1.5248	-
0.0362	3620	1.5527	-
0.0363	3630	1.5524	-
0.0364	3640	1.531	-
0.0365	3650	1.5358	-
0.0366	3660	1.5495	-
0.0367	3670	1.5415	-
0.0368	3680	1.5612	-
0.0369	3690	1.5284	-
0.037	3700	1.5384	0.5264
0.0371	3710	1.5432	-
0.0372	3720	1.538	-
0.0373	3730	1.5139	-
0.0374	3740	1.5401	-
0.0375	3750	1.5535	-
0.0376	3760	1.5559	-
0.0377	3770	1.5608	-
0.0378	3780	1.5484	-
0.0379	3790	1.557	-
0.038	3800	1.5436	0.5365
0.0381	3810	1.5461	-
0.0382	3820	1.548	-
0.0383	3830	1.5419	-
0.0384	3840	1.5495	-
0.0385	3850	1.5222	-
0.0386	3860	1.5303	-
0.0387	3870	1.5645	-
0.0388	3880	1.5479	-
0.0389	3890	1.5324	-
0.039	3900	1.5456	0.5349
0.0391	3910	1.5304	-
0.0392	3920	1.5494	-
0.0393	3930	1.557	-
0.0394	3940	1.5526	-
0.0395	3950	1.5351	-
0.0396	3960	1.5275	-
0.0397	3970	1.5411	-
0.0398	3980	1.5199	-
0.0399	3990	1.538	-
0.04	4000	1.5346	0.5361
0.0401	4010	1.5543	-
0.0402	4020	1.5451	-
0.0403	4030	1.541	-
0.0404	4040	1.5287	-
0.0405	4050	1.5391	-
0.0406	4060	1.5198	-
0.0407	4070	1.5499	-
0.0408	4080	1.538	-
0.0409	4090	1.5429	-
0.041	4100	1.5415	0.5384
0.0411	4110	1.5414	-
0.0412	4120	1.5286	-
0.0413	4130	1.5336	-
0.0414	4140	1.5199	-
0.0415	4150	1.555	-
0.0416	4160	1.5299	-
0.0417	4170	1.5189	-
0.0418	4180	1.5223	-
0.0419	4190	1.5157	-
0.042	4200	1.5273	0.5365
0.0421	4210	1.5244	-
0.0422	4220	1.5154	-
0.0423	4230	1.5447	-
0.0424	4240	1.5345	-
0.0425	4250	1.5341	-
0.0426	4260	1.511	-
0.0427	4270	1.5194	-
0.0428	4280	1.5553	-
0.0429	4290	1.526	-
0.043	4300	1.5117	0.5317
0.0431	4310	1.5275	-
0.0432	4320	1.539	-
0.0433	4330	1.5391	-
0.0434	4340	1.5203	-
0.0435	4350	1.5251	-
0.0436	4360	1.5146	-
0.0437	4370	1.5524	-
0.0438	4380	1.5178	-
0.0439	4390	1.5381	-
0.044	4400	1.532	0.5366
0.0441	4410	1.5376	-
0.0442	4420	1.5313	-
0.0443	4430	1.5304	-
0.0444	4440	1.5345	-
0.0445	4450	1.5139	-
0.0446	4460	1.5263	-
0.0447	4470	1.5085	-
0.0448	4480	1.5135	-
0.0449	4490	1.5282	-
0.045	4500	1.5154	0.5338
0.0451	4510	1.5285	-
0.0452	4520	1.5325	-
0.0453	4530	1.5272	-
0.0454	4540	1.5212	-
0.0455	4550	1.509	-
0.0456	4560	1.5197	-
0.0457	4570	1.542	-
0.0458	4580	1.5315	-
0.0459	4590	1.4863	-
0.046	4600	1.5197	0.5357
0.0461	4610	1.5321	-
0.0462	4620	1.5214	-
0.0463	4630	1.5183	-
0.0464	4640	1.5184	-
0.0465	4650	1.5093	-
0.0466	4660	1.4965	-
0.0467	4670	1.5134	-
0.0468	4680	1.5006	-
0.0469	4690	1.5316	-
0.047	4700	1.511	0.5365
0.0471	4710	1.5317	-
0.0472	4720	1.5109	-
0.0473	4730	1.5409	-
0.0474	4740	1.5194	-
0.0475	4750	1.5031	-
0.0476	4760	1.5349	-
0.0477	4770	1.5151	-
0.0478	4780	1.5113	-
0.0479	4790	1.519	-
0.048	4800	1.502	0.5289
0.0481	4810	1.5291	-
0.0482	4820	1.5124	-
0.0483	4830	1.5064	-
0.0484	4840	1.519	-
0.0485	4850	1.5157	-
0.0486	4860	1.513	-
0.0487	4870	1.4979	-
0.0488	4880	1.5048	-
0.0489	4890	1.4867	-
0.049	4900	1.5199	0.5402
0.0491	4910	1.5196	-
0.0492	4920	1.5267	-
0.0493	4930	1.4868	-
0.0494	4940	1.5305	-
0.0495	4950	1.5289	-
0.0496	4960	1.5117	-
0.0497	4970	1.5105	-
0.0498	4980	1.5063	-
0.0499	4990	1.5224	-
0.05	5000	1.4837	0.5356
0.0501	5010	1.5313	-
0.0502	5020	1.5189	-
0.0503	5030	1.4985	-
0.0504	5040	1.4909	-
0.0505	5050	1.5181	-
0.0506	5060	1.4973	-
0.0507	5070	1.5351	-
0.0508	5080	1.5303	-
0.0509	5090	1.5066	-
0.051	5100	1.5061	0.5376
0.0511	5110	1.5115	-
0.0512	5120	1.5103	-
0.0513	5130	1.5114	-
0.0514	5140	1.5203	-
0.0515	5150	1.5254	-
0.0516	5160	1.5163	-
0.0517	5170	1.4856	-
0.0518	5180	1.524	-
0.0519	5190	1.5323	-
0.052	5200	1.5157	0.5396
0.0521	5210	1.4932	-
0.0522	5220	1.5132	-
0.0523	5230	1.5004	-
0.0524	5240	1.5061	-
0.0525	5250	1.4832	-
0.0526	5260	1.529	-
0.0527	5270	1.5053	-
0.0528	5280	1.5093	-
0.0529	5290	1.4882	-
0.053	5300	1.506	0.5381
0.0531	5310	1.5227	-
0.0532	5320	1.535	-
0.0533	5330	1.4911	-
0.0534	5340	1.5182	-
0.0535	5350	1.5181	-
0.0536	5360	1.4936	-
0.0537	5370	1.4962	-
0.0538	5380	1.5083	-
0.0539	5390	1.4875	-
0.054	5400	1.4968	0.5435
0.0541	5410	1.5252	-
0.0542	5420	1.5126	-
0.0543	5430	1.5135	-
0.0544	5440	1.5129	-
0.0545	5450	1.4987	-
0.0546	5460	1.5031	-
0.0547	5470	1.5212	-
0.0548	5480	1.5244	-
0.0549	5490	1.5156	-
0.055	5500	1.4782	0.5417
0.0551	5510	1.5206	-
0.0552	5520	1.5096	-
0.0553	5530	1.5113	-
0.0554	5540	1.5079	-
0.0555	5550	1.5081	-
0.0556	5560	1.5208	-
0.0557	5570	1.4893	-
0.0558	5580	1.5032	-
0.0559	5590	1.4956	-
0.056	5600	1.529	0.5364
0.0561	5610	1.5204	-
0.0562	5620	1.4714	-
0.0563	5630	1.4958	-
0.0564	5640	1.496	-
0.0565	5650	1.5116	-
0.0566	5660	1.5061	-
0.0567	5670	1.4922	-
0.0568	5680	1.5214	-
0.0569	5690	1.5205	-
0.057	5700	1.5045	0.5423
0.0571	5710	1.4884	-
0.0572	5720	1.5151	-
0.0573	5730	1.494	-
0.0574	5740	1.4974	-
0.0575	5750	1.4691	-
0.0576	5760	1.5139	-
0.0577	5770	1.5019	-
0.0578	5780	1.5264	-
0.0579	5790	1.4845	-
0.058	5800	1.4967	0.5402
0.0581	5810	1.5033	-
0.0582	5820	1.5003	-
0.0583	5830	1.5008	-
0.0584	5840	1.4808	-
0.0585	5850	1.4833	-
0.0586	5860	1.5167	-
0.0587	5870	1.5093	-
0.0588	5880	1.522	-
0.0589	5890	1.4933	-
0.059	5900	1.4948	0.5442
0.0591	5910	1.4988	-
0.0592	5920	1.4881	-
0.0593	5930	1.5188	-
0.0594	5940	1.4747	-
0.0595	5950	1.4982	-
0.0596	5960	1.4678	-
0.0597	5970	1.482	-
0.0598	5980	1.5238	-
0.0599	5990	1.5182	-
0.06	6000	1.4929	0.5387
0.0601	6010	1.4804	-
0.0602	6020	1.4982	-
0.0603	6030	1.4932	-
0.0604	6040	1.4962	-
0.0605	6050	1.5051	-
0.0606	6060	1.5066	-
0.0607	6070	1.4766	-
0.0608	6080	1.4815	-
0.0609	6090	1.488	-
0.061	6100	1.4994	0.5409
0.0611	6110	1.4711	-
0.0612	6120	1.4799	-
0.0613	6130	1.4951	-
0.0614	6140	1.5054	-
0.0615	6150	1.5174	-
0.0616	6160	1.4866	-
0.0617	6170	1.4981	-
0.0618	6180	1.5147	-
0.0619	6190	1.493	-
0.062	6200	1.4789	0.5435

Framework Versions

Python: 3.12.12
Sentence Transformers: 5.2.0
Transformers: 4.57.3
PyTorch: 2.9.0+cu128
Accelerate: 1.12.0
Datasets: 4.4.2
Tokenizers: 0.22.1

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",
}

CachedMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}