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_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, 0.8640, 0.8773],
#         [0.8640, 1.0000, 0.7820],
#         [0.8773, 0.7820, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: validation_retrieval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.57
cosine_accuracy@3	0.82
cosine_accuracy@5	0.89
cosine_accuracy@10	0.94
cosine_precision@1	0.57
cosine_precision@3	0.4767
cosine_precision@5	0.446
cosine_precision@10	0.327
cosine_recall@1	0.1145
cosine_recall@3	0.2377
cosine_recall@5	0.3327
cosine_recall@10	0.4088
cosine_ndcg@10	0.508
cosine_ndcg@100	0.5643
cosine_mrr@10	0.7045
cosine_mrr@100	0.7066
cosine_map@100	0.3787

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	5.3032	-
0.0001	10	3.7433	-
0.0002	20	2.7632	-
0.0003	30	2.4609	-
0.0004	40	2.303	-
0.0005	50	2.2197	-
0.0006	60	2.1855	-
0.0007	70	2.1517	-
0.0008	80	2.1111	-
0.0009	90	2.0876	-
0.001	100	2.0613	0.4992
0.0011	110	2.0348	-
0.0012	120	2.0209	-
0.0013	130	2.0293	-
0.0014	140	2.0281	-
0.0015	150	2.0019	-
0.0016	160	1.9692	-
0.0017	170	1.9849	-
0.0018	180	1.9494	-
0.0019	190	1.9458	-
0.002	200	1.9264	0.5104
0.0021	210	1.9644	-
0.0022	220	1.9315	-
0.0023	230	1.9324	-
0.0024	240	1.8965	-
0.0025	250	1.9287	-
0.0026	260	1.9246	-
0.0027	270	1.9087	-
0.0028	280	1.9052	-
0.0029	290	1.8969	-
0.003	300	1.8971	0.5164
0.0031	310	1.8896	-
0.0032	320	1.8897	-
0.0033	330	1.8646	-
0.0034	340	1.8825	-
0.0035	350	1.8599	-
0.0036	360	1.8583	-
0.0037	370	1.8649	-
0.0038	380	1.8647	-
0.0039	390	1.8759	-
0.004	400	1.8197	0.5253
0.0041	410	1.846	-
0.0042	420	1.841	-
0.0043	430	1.8319	-
0.0044	440	1.835	-
0.0045	450	1.807	-
0.0046	460	1.8406	-
0.0047	470	1.8344	-
0.0048	480	1.8003	-
0.0049	490	1.8155	-
0.005	500	1.8242	0.5266
0.0051	510	1.8014	-
0.0052	520	1.8026	-
0.0053	530	1.8042	-
0.0054	540	1.8372	-
0.0055	550	1.8054	-
0.0056	560	1.8093	-
0.0057	570	1.7814	-
0.0058	580	1.7875	-
0.0059	590	1.7844	-
0.006	600	1.7789	0.5330
0.0061	610	1.7947	-
0.0062	620	1.8084	-
0.0063	630	1.7806	-
0.0064	640	1.772	-
0.0065	650	1.7948	-
0.0066	660	1.7648	-
0.0067	670	1.7801	-
0.0068	680	1.7801	-
0.0069	690	1.7696	-
0.007	700	1.7848	0.5457
0.0071	710	1.774	-
0.0072	720	1.7619	-
0.0073	730	1.7422	-
0.0074	740	1.7594	-
0.0075	750	1.7225	-
0.0076	760	1.7601	-
0.0077	770	1.7432	-
0.0078	780	1.7627	-
0.0079	790	1.749	-
0.008	800	1.7361	0.5455
0.0081	810	1.7275	-
0.0082	820	1.7391	-
0.0083	830	1.7403	-
0.0084	840	1.736	-
0.0085	850	1.7297	-
0.0086	860	1.7376	-
0.0087	870	1.7242	-
0.0088	880	1.7231	-
0.0089	890	1.729	-
0.009	900	1.7515	0.5473
0.0091	910	1.7269	-
0.0092	920	1.6863	-
0.0093	930	1.7164	-
0.0094	940	1.7347	-
0.0095	950	1.7439	-
0.0096	960	1.7102	-
0.0097	970	1.7129	-
0.0098	980	1.7185	-
0.0099	990	1.7131	-
0.01	1000	1.7309	0.5527
0.0101	1010	1.7055	-
0.0102	1020	1.7106	-
0.0103	1030	1.7089	-
0.0104	1040	1.7058	-
0.0105	1050	1.6984	-
0.0106	1060	1.69	-
0.0107	1070	1.7189	-
0.0108	1080	1.7147	-
0.0109	1090	1.7237	-
0.011	1100	1.6781	0.5567
0.0111	1110	1.6788	-
0.0112	1120	1.6928	-
0.0113	1130	1.7146	-
0.0114	1140	1.6983	-
0.0115	1150	1.7014	-
0.0116	1160	1.6888	-
0.0117	1170	1.6668	-
0.0118	1180	1.6785	-
0.0119	1190	1.6853	-
0.012	1200	1.7077	0.5459
0.0121	1210	1.676	-
0.0122	1220	1.6749	-
0.0123	1230	1.6815	-
0.0124	1240	1.6823	-
0.0125	1250	1.6751	-
0.0126	1260	1.6942	-
0.0127	1270	1.6597	-
0.0128	1280	1.6685	-
0.0129	1290	1.6873	-
0.013	1300	1.6779	0.5526
0.0131	1310	1.6676	-
0.0132	1320	1.6721	-
0.0133	1330	1.6713	-
0.0134	1340	1.6618	-
0.0135	1350	1.6387	-
0.0136	1360	1.6951	-
0.0137	1370	1.6669	-
0.0138	1380	1.6477	-
0.0139	1390	1.6856	-
0.014	1400	1.6687	0.5528
0.0141	1410	1.6578	-
0.0142	1420	1.6588	-
0.0143	1430	1.6552	-
0.0144	1440	1.6643	-
0.0145	1450	1.6543	-
0.0146	1460	1.6851	-
0.0147	1470	1.6547	-
0.0148	1480	1.6744	-
0.0149	1490	1.6694	-
0.015	1500	1.6795	0.5537
0.0151	1510	1.656	-
0.0152	1520	1.6425	-
0.0153	1530	1.6545	-
0.0154	1540	1.614	-
0.0155	1550	1.6554	-
0.0156	1560	1.6542	-
0.0157	1570	1.6676	-
0.0158	1580	1.6615	-
0.0159	1590	1.6374	-
0.016	1600	1.6451	0.5613
0.0161	1610	1.6258	-
0.0162	1620	1.6504	-
0.0163	1630	1.6254	-
0.0164	1640	1.6257	-
0.0165	1650	1.6392	-
0.0166	1660	1.6365	-
0.0167	1670	1.6407	-
0.0168	1680	1.6313	-
0.0169	1690	1.6458	-
0.017	1700	1.6405	0.5526
0.0171	1710	1.6431	-
0.0172	1720	1.6262	-
0.0173	1730	1.6434	-
0.0174	1740	1.6404	-
0.0175	1750	1.6418	-
0.0176	1760	1.6176	-
0.0177	1770	1.6282	-
0.0178	1780	1.6228	-
0.0179	1790	1.656	-
0.018	1800	1.6392	0.5499
0.0181	1810	1.6307	-
0.0182	1820	1.6147	-
0.0183	1830	1.6225	-
0.0184	1840	1.6387	-
0.0185	1850	1.6173	-
0.0186	1860	1.6535	-
0.0187	1870	1.6339	-
0.0188	1880	1.6215	-
0.0189	1890	1.6048	-
0.019	1900	1.6278	0.5527
0.0191	1910	1.6359	-
0.0192	1920	1.6142	-
0.0193	1930	1.6354	-
0.0194	1940	1.6341	-
0.0195	1950	1.6352	-
0.0196	1960	1.6223	-
0.0197	1970	1.6208	-
0.0198	1980	1.6151	-
0.0199	1990	1.5815	-
0.02	2000	1.6159	0.5573
0.0201	2010	1.6229	-
0.0202	2020	1.6156	-
0.0203	2030	1.6051	-
0.0204	2040	1.6411	-
0.0205	2050	1.6339	-
0.0206	2060	1.6241	-
0.0207	2070	1.6014	-
0.0208	2080	1.5942	-
0.0209	2090	1.611	-
0.021	2100	1.6065	0.5563
0.0211	2110	1.6208	-
0.0212	2120	1.6239	-
0.0213	2130	1.6066	-
0.0214	2140	1.5936	-
0.0215	2150	1.6008	-
0.0216	2160	1.6239	-
0.0217	2170	1.6116	-
0.0218	2180	1.6128	-
0.0219	2190	1.5819	-
0.022	2200	1.5915	0.5547
0.0221	2210	1.6164	-
0.0222	2220	1.6141	-
0.0223	2230	1.6296	-
0.0224	2240	1.6026	-
0.0225	2250	1.5958	-
0.0226	2260	1.6009	-
0.0227	2270	1.6336	-
0.0228	2280	1.6231	-
0.0229	2290	1.6163	-
0.023	2300	1.5811	0.5626
0.0231	2310	1.5951	-
0.0232	2320	1.5989	-
0.0233	2330	1.6056	-
0.0234	2340	1.5808	-
0.0235	2350	1.5741	-
0.0236	2360	1.5928	-
0.0237	2370	1.5921	-
0.0238	2380	1.6032	-
0.0239	2390	1.5779	-
0.024	2400	1.609	0.5637
0.0241	2410	1.5771	-
0.0242	2420	1.5902	-
0.0243	2430	1.5971	-
0.0244	2440	1.5969	-
0.0245	2450	1.6058	-
0.0246	2460	1.6161	-
0.0247	2470	1.5709	-
0.0248	2480	1.5814	-
0.0249	2490	1.5866	-
0.025	2500	1.5692	0.5642
0.0251	2510	1.584	-
0.0252	2520	1.5899	-
0.0253	2530	1.614	-
0.0254	2540	1.5966	-
0.0255	2550	1.5838	-
0.0256	2560	1.5969	-
0.0257	2570	1.5789	-
0.0258	2580	1.5938	-
0.0259	2590	1.5836	-
0.026	2600	1.579	0.5640
0.0261	2610	1.5978	-
0.0262	2620	1.5783	-
0.0263	2630	1.5842	-
0.0264	2640	1.6001	-
0.0265	2650	1.5798	-
0.0266	2660	1.6003	-
0.0267	2670	1.5868	-
0.0268	2680	1.603	-
0.0269	2690	1.5789	-
0.027	2700	1.5724	0.5674
0.0271	2710	1.5718	-
0.0272	2720	1.5771	-
0.0273	2730	1.5954	-
0.0274	2740	1.5687	-
0.0275	2750	1.5897	-
0.0276	2760	1.5533	-
0.0277	2770	1.5799	-
0.0278	2780	1.5741	-
0.0279	2790	1.6096	-
0.028	2800	1.5863	0.5568
0.0281	2810	1.6004	-
0.0282	2820	1.569	-
0.0283	2830	1.5757	-
0.0284	2840	1.5597	-
0.0285	2850	1.5935	-
0.0286	2860	1.5673	-
0.0287	2870	1.5725	-
0.0288	2880	1.5899	-
0.0289	2890	1.5683	-
0.029	2900	1.5519	0.5702
0.0291	2910	1.559	-
0.0292	2920	1.5692	-
0.0293	2930	1.5792	-
0.0294	2940	1.5704	-
0.0295	2950	1.5717	-
0.0296	2960	1.5535	-
0.0297	2970	1.553	-
0.0298	2980	1.5629	-
0.0299	2990	1.5636	-
0.03	3000	1.5715	0.5681
0.0301	3010	1.5538	-
0.0302	3020	1.5803	-
0.0303	3030	1.5535	-
0.0304	3040	1.5674	-
0.0305	3050	1.5465	-
0.0306	3060	1.5682	-
0.0307	3070	1.5855	-
0.0308	3080	1.559	-
0.0309	3090	1.559	-
0.031	3100	1.5773	0.5707
0.0311	3110	1.5693	-
0.0312	3120	1.5643	-
0.0313	3130	1.5586	-
0.0314	3140	1.5453	-
0.0315	3150	1.5799	-
0.0316	3160	1.5532	-
0.0317	3170	1.5459	-
0.0318	3180	1.5541	-
0.0319	3190	1.5789	-
0.032	3200	1.5331	0.5595
0.0321	3210	1.5521	-
0.0322	3220	1.5553	-
0.0323	3230	1.5675	-
0.0324	3240	1.551	-
0.0325	3250	1.5753	-
0.0326	3260	1.5625	-
0.0327	3270	1.5782	-
0.0328	3280	1.5588	-
0.0329	3290	1.5795	-
0.033	3300	1.5529	0.5654
0.0331	3310	1.5581	-
0.0332	3320	1.5828	-
0.0333	3330	1.5628	-
0.0334	3340	1.5614	-
0.0335	3350	1.5645	-
0.0336	3360	1.5405	-
0.0337	3370	1.5743	-
0.0338	3380	1.5393	-
0.0339	3390	1.5547	-
0.034	3400	1.5403	0.5616
0.0341	3410	1.5627	-
0.0342	3420	1.5638	-
0.0343	3430	1.5664	-
0.0344	3440	1.5345	-
0.0345	3450	1.5546	-
0.0346	3460	1.5581	-
0.0347	3470	1.5614	-
0.0348	3480	1.558	-
0.0349	3490	1.5451	-
0.035	3500	1.5491	0.5581
0.0351	3510	1.5357	-
0.0352	3520	1.5578	-
0.0353	3530	1.5433	-
0.0354	3540	1.5343	-
0.0355	3550	1.5558	-
0.0356	3560	1.5711	-
0.0357	3570	1.5458	-
0.0358	3580	1.5356	-
0.0359	3590	1.559	-
0.036	3600	1.5338	0.5598
0.0361	3610	1.5532	-
0.0362	3620	1.5346	-
0.0363	3630	1.5558	-
0.0364	3640	1.539	-
0.0365	3650	1.538	-
0.0366	3660	1.5638	-
0.0367	3670	1.5666	-
0.0368	3680	1.5662	-
0.0369	3690	1.5432	-
0.037	3700	1.5345	0.5680
0.0371	3710	1.5524	-
0.0372	3720	1.5617	-
0.0373	3730	1.5261	-
0.0374	3740	1.5502	-
0.0375	3750	1.5452	-
0.0376	3760	1.5566	-
0.0377	3770	1.5457	-
0.0378	3780	1.5307	-
0.0379	3790	1.5331	-
0.038	3800	1.5294	0.5578
0.0381	3810	1.5389	-
0.0382	3820	1.5379	-
0.0383	3830	1.5578	-
0.0384	3840	1.5259	-
0.0385	3850	1.5308	-
0.0386	3860	1.5461	-
0.0387	3870	1.5197	-
0.0388	3880	1.5332	-
0.0389	3890	1.5642	-
0.039	3900	1.5256	0.5625
0.0391	3910	1.5608	-
0.0392	3920	1.5567	-
0.0393	3930	1.5278	-
0.0394	3940	1.5404	-
0.0395	3950	1.5367	-
0.0396	3960	1.5186	-
0.0397	3970	1.5437	-
0.0398	3980	1.5459	-
0.0399	3990	1.5536	-
0.04	4000	1.548	0.5642
0.0401	4010	1.5407	-
0.0402	4020	1.5235	-
0.0403	4030	1.526	-
0.0404	4040	1.5184	-
0.0405	4050	1.5232	-
0.0406	4060	1.5215	-
0.0407	4070	1.5202	-
0.0408	4080	1.5325	-
0.0409	4090	1.5317	-
0.041	4100	1.5326	0.5689
0.0411	4110	1.5083	-
0.0412	4120	1.5158	-
0.0413	4130	1.5321	-
0.0414	4140	1.5383	-
0.0415	4150	1.5432	-
0.0416	4160	1.503	-
0.0417	4170	1.5374	-
0.0418	4180	1.5166	-
0.0419	4190	1.5462	-
0.042	4200	1.5175	0.5650
0.0421	4210	1.5348	-
0.0422	4220	1.5613	-
0.0423	4230	1.521	-
0.0424	4240	1.5377	-
0.0425	4250	1.5163	-
0.0426	4260	1.5354	-
0.0427	4270	1.5181	-
0.0428	4280	1.5381	-
0.0429	4290	1.5311	-
0.043	4300	1.5074	0.5688
0.0431	4310	1.5162	-
0.0432	4320	1.5051	-
0.0433	4330	1.5171	-
0.0434	4340	1.5283	-
0.0435	4350	1.5171	-
0.0436	4360	1.5377	-
0.0437	4370	1.5197	-
0.0438	4380	1.513	-
0.0439	4390	1.5418	-
0.044	4400	1.5135	0.5644
0.0441	4410	1.522	-
0.0442	4420	1.5286	-
0.0443	4430	1.5328	-
0.0444	4440	1.5354	-
0.0445	4450	1.5252	-
0.0446	4460	1.5127	-
0.0447	4470	1.5116	-
0.0448	4480	1.5237	-
0.0449	4490	1.5265	-
0.045	4500	1.5298	0.5649
0.0451	4510	1.5349	-
0.0452	4520	1.4997	-
0.0453	4530	1.4947	-
0.0454	4540	1.5186	-
0.0455	4550	1.487	-
0.0456	4560	1.5088	-
0.0457	4570	1.5422	-
0.0458	4580	1.4962	-
0.0459	4590	1.5193	-
0.046	4600	1.5306	0.5608
0.0461	4610	1.536	-
0.0462	4620	1.5334	-
0.0463	4630	1.5598	-
0.0464	4640	1.5223	-
0.0465	4650	1.5223	-
0.0466	4660	1.5277	-
0.0467	4670	1.5381	-
0.0468	4680	1.5416	-
0.0469	4690	1.5056	-
0.047	4700	1.5077	0.5655
0.0471	4710	1.5045	-
0.0472	4720	1.5135	-
0.0473	4730	1.5284	-
0.0474	4740	1.5331	-
0.0475	4750	1.5194	-
0.0476	4760	1.5286	-
0.0477	4770	1.536	-
0.0478	4780	1.4984	-
0.0479	4790	1.5086	-
0.048	4800	1.5137	0.5703
0.0481	4810	1.5421	-
0.0482	4820	1.5131	-
0.0483	4830	1.5084	-
0.0484	4840	1.5006	-
0.0485	4850	1.5141	-
0.0486	4860	1.503	-
0.0487	4870	1.511	-
0.0488	4880	1.5175	-
0.0489	4890	1.5088	-
0.049	4900	1.5019	0.5711
0.0491	4910	1.5359	-
0.0492	4920	1.5218	-
0.0493	4930	1.5043	-
0.0494	4940	1.5059	-
0.0495	4950	1.4943	-
0.0496	4960	1.5269	-
0.0497	4970	1.517	-
0.0498	4980	1.5135	-
0.0499	4990	1.5204	-
0.05	5000	1.4983	0.5700
0.0501	5010	1.5271	-
0.0502	5020	1.4929	-
0.0503	5030	1.4947	-
0.0504	5040	1.4883	-
0.0505	5050	1.523	-
0.0506	5060	1.5092	-
0.0507	5070	1.5262	-
0.0508	5080	1.4859	-
0.0509	5090	1.5059	-
0.051	5100	1.5293	0.5677
0.0511	5110	1.4962	-
0.0512	5120	1.5192	-
0.0513	5130	1.5115	-
0.0514	5140	1.5152	-
0.0515	5150	1.4948	-
0.0516	5160	1.5376	-
0.0517	5170	1.5015	-
0.0518	5180	1.5119	-
0.0519	5190	1.4926	-
0.052	5200	1.5235	0.5663
0.0521	5210	1.5158	-
0.0522	5220	1.5072	-
0.0523	5230	1.5264	-
0.0524	5240	1.5026	-
0.0525	5250	1.5042	-
0.0526	5260	1.5096	-
0.0527	5270	1.5022	-
0.0528	5280	1.5038	-
0.0529	5290	1.4903	-
0.053	5300	1.5284	0.5684
0.0531	5310	1.5009	-
0.0532	5320	1.505	-
0.0533	5330	1.5288	-
0.0534	5340	1.501	-
0.0535	5350	1.5143	-
0.0536	5360	1.5071	-
0.0537	5370	1.4976	-
0.0538	5380	1.5092	-
0.0539	5390	1.5082	-
0.054	5400	1.5056	0.5716
0.0541	5410	1.4934	-
0.0542	5420	1.5159	-
0.0543	5430	1.5059	-
0.0544	5440	1.4937	-
0.0545	5450	1.5223	-
0.0546	5460	1.4989	-
0.0547	5470	1.5149	-
0.0548	5480	1.5003	-
0.0549	5490	1.521	-
0.055	5500	1.4959	0.5779
0.0551	5510	1.5074	-
0.0552	5520	1.5071	-
0.0553	5530	1.5173	-
0.0554	5540	1.5111	-
0.0555	5550	1.5017	-
0.0556	5560	1.5296	-
0.0557	5570	1.5147	-
0.0558	5580	1.524	-
0.0559	5590	1.4936	-
0.056	5600	1.5111	0.5684
0.0561	5610	1.5147	-
0.0562	5620	1.5002	-
0.0563	5630	1.5048	-
0.0564	5640	1.5093	-
0.0565	5650	1.5093	-
0.0566	5660	1.4795	-
0.0567	5670	1.5149	-
0.0568	5680	1.4881	-
0.0569	5690	1.4986	-
0.057	5700	1.4929	0.5692
0.0571	5710	1.5186	-
0.0572	5720	1.4938	-
0.0573	5730	1.4943	-
0.0574	5740	1.4926	-
0.0575	5750	1.4672	-
0.0576	5760	1.5036	-
0.0577	5770	1.511	-
0.0578	5780	1.4892	-
0.0579	5790	1.4983	-
0.058	5800	1.4914	0.5704
0.0581	5810	1.4883	-
0.0582	5820	1.5052	-
0.0583	5830	1.5066	-
0.0584	5840	1.4904	-
0.0585	5850	1.5114	-
0.0586	5860	1.4984	-
0.0587	5870	1.4827	-
0.0588	5880	1.4676	-
0.0589	5890	1.514	-
0.059	5900	1.509	0.5688
0.0591	5910	1.5094	-
0.0592	5920	1.4902	-
0.0593	5930	1.4849	-
0.0594	5940	1.5159	-
0.0595	5950	1.5012	-
0.0596	5960	1.5068	-
0.0597	5970	1.5054	-
0.0598	5980	1.4722	-
0.0599	5990	1.4975	-
0.06	6000	1.4843	0.5623
0.0601	6010	1.4726	-
0.0602	6020	1.517	-
0.0603	6030	1.4957	-
0.0604	6040	1.508	-
0.0605	6050	1.5113	-
0.0606	6060	1.4903	-
0.0607	6070	1.4761	-
0.0608	6080	1.5226	-
0.0609	6090	1.5228	-
0.061	6100	1.4836	0.5643
0.0611	6110	1.4926	-
0.0612	6120	1.4968	-
0.0613	6130	1.4954	-
0.0614	6140	1.5209	-
0.0615	6150	1.4857	-
0.0616	6160	1.4881	-
0.0617	6170	1.504	-
0.0618	6180	1.464	-
0.0619	6190	1.5003	-
0.062	6200	1.4858	0.5643

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}
}