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_ac1_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, 0.8605, 0.8626],
#         [0.8605, 1.0000, 0.7646],
#         [0.8626, 0.7646, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: validation_retrieval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.62
cosine_accuracy@3	0.8
cosine_accuracy@5	0.85
cosine_accuracy@10	0.92
cosine_precision@1	0.62
cosine_precision@3	0.5
cosine_precision@5	0.436
cosine_precision@10	0.336
cosine_recall@1	0.1222
cosine_recall@3	0.2575
cosine_recall@5	0.3172
cosine_recall@10	0.4205
cosine_ndcg@10	0.5236
cosine_ndcg@100	0.579
cosine_mrr@10	0.7175
cosine_mrr@100	0.7209
cosine_map@100	0.3916

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	5.3619	-
0.0001	10	3.792	-
0.0002	20	2.7885	-
0.0003	30	2.4762	-
0.0004	40	2.3427	-
0.0005	50	2.2636	-
0.0006	60	2.2051	-
0.0007	70	2.1559	-
0.0008	80	2.1319	-
0.0009	90	2.1058	-
0.001	100	2.0827	0.5286
0.0011	110	2.0716	-
0.0012	120	2.0793	-
0.0013	130	2.0351	-
0.0014	140	2.0021	-
0.0015	150	2.0188	-
0.0016	160	2.004	-
0.0017	170	2.0122	-
0.0018	180	2.001	-
0.0019	190	1.9768	-
0.002	200	1.9975	0.5425
0.0021	210	1.9717	-
0.0022	220	1.9549	-
0.0023	230	1.9626	-
0.0024	240	1.927	-
0.0025	250	1.9438	-
0.0026	260	1.9322	-
0.0027	270	1.9176	-
0.0028	280	1.9225	-
0.0029	290	1.9225	-
0.003	300	1.8604	0.5460
0.0031	310	1.9088	-
0.0032	320	1.8923	-
0.0033	330	1.8864	-
0.0034	340	1.8888	-
0.0035	350	1.9083	-
0.0036	360	1.8757	-
0.0037	370	1.8743	-
0.0038	380	1.8348	-
0.0039	390	1.8616	-
0.004	400	1.865	0.5477
0.0041	410	1.8715	-
0.0042	420	1.8706	-
0.0043	430	1.853	-
0.0044	440	1.8556	-
0.0045	450	1.825	-
0.0046	460	1.8544	-
0.0047	470	1.8383	-
0.0048	480	1.839	-
0.0049	490	1.8289	-
0.005	500	1.8222	0.5500
0.0051	510	1.8373	-
0.0052	520	1.8183	-
0.0053	530	1.8254	-
0.0054	540	1.7999	-
0.0055	550	1.8108	-
0.0056	560	1.8101	-
0.0057	570	1.8243	-
0.0058	580	1.8348	-
0.0059	590	1.7991	-
0.006	600	1.7865	0.5617
0.0061	610	1.8015	-
0.0062	620	1.7859	-
0.0063	630	1.7863	-
0.0064	640	1.7965	-
0.0065	650	1.8217	-
0.0066	660	1.7747	-
0.0067	670	1.802	-
0.0068	680	1.7932	-
0.0069	690	1.7734	-
0.007	700	1.776	0.5635
0.0071	710	1.7798	-
0.0072	720	1.7702	-
0.0073	730	1.7753	-
0.0074	740	1.7488	-
0.0075	750	1.7676	-
0.0076	760	1.7855	-
0.0077	770	1.7672	-
0.0078	780	1.7847	-
0.0079	790	1.7669	-
0.008	800	1.7564	0.5608
0.0081	810	1.7619	-
0.0082	820	1.7366	-
0.0083	830	1.7396	-
0.0084	840	1.751	-
0.0085	850	1.7604	-
0.0086	860	1.768	-
0.0087	870	1.7424	-
0.0088	880	1.7533	-
0.0089	890	1.7156	-
0.009	900	1.7449	0.5620
0.0091	910	1.7198	-
0.0092	920	1.734	-
0.0093	930	1.729	-
0.0094	940	1.7162	-
0.0095	950	1.7283	-
0.0096	960	1.7101	-
0.0097	970	1.7252	-
0.0098	980	1.7391	-
0.0099	990	1.7371	-
0.01	1000	1.7161	0.5685
0.0101	1010	1.7148	-
0.0102	1020	1.7314	-
0.0103	1030	1.7352	-
0.0104	1040	1.7156	-
0.0105	1050	1.7308	-
0.0106	1060	1.7348	-
0.0107	1070	1.7334	-
0.0108	1080	1.7177	-
0.0109	1090	1.7029	-
0.011	1100	1.7357	0.5819
0.0111	1110	1.7081	-
0.0112	1120	1.6928	-
0.0113	1130	1.7355	-
0.0114	1140	1.7071	-
0.0115	1150	1.706	-
0.0116	1160	1.7074	-
0.0117	1170	1.6842	-
0.0118	1180	1.6944	-
0.0119	1190	1.7054	-
0.012	1200	1.7458	0.5704
0.0121	1210	1.7028	-
0.0122	1220	1.7026	-
0.0123	1230	1.7061	-
0.0124	1240	1.6791	-
0.0125	1250	1.6996	-
0.0126	1260	1.7036	-
0.0127	1270	1.7152	-
0.0128	1280	1.7158	-
0.0129	1290	1.6871	-
0.013	1300	1.6842	0.5736
0.0131	1310	1.6887	-
0.0132	1320	1.6747	-
0.0133	1330	1.6982	-
0.0134	1340	1.6845	-
0.0135	1350	1.6801	-
0.0136	1360	1.6926	-
0.0137	1370	1.6747	-
0.0138	1380	1.6719	-
0.0139	1390	1.6839	-
0.014	1400	1.6722	0.5754
0.0141	1410	1.6779	-
0.0142	1420	1.6827	-
0.0143	1430	1.6728	-
0.0144	1440	1.6871	-
0.0145	1450	1.6958	-
0.0146	1460	1.6735	-
0.0147	1470	1.6654	-
0.0148	1480	1.6831	-
0.0149	1490	1.6811	-
0.015	1500	1.673	0.5725
0.0151	1510	1.6589	-
0.0152	1520	1.6465	-
0.0153	1530	1.6808	-
0.0154	1540	1.6528	-
0.0155	1550	1.6884	-
0.0156	1560	1.6496	-
0.0157	1570	1.6469	-
0.0158	1580	1.6774	-
0.0159	1590	1.6869	-
0.016	1600	1.6426	0.5710
0.0161	1610	1.6656	-
0.0162	1620	1.6735	-
0.0163	1630	1.6886	-
0.0164	1640	1.6664	-
0.0165	1650	1.6578	-
0.0166	1660	1.6656	-
0.0167	1670	1.6729	-
0.0168	1680	1.665	-
0.0169	1690	1.6612	-
0.017	1700	1.6599	0.5691
0.0171	1710	1.637	-
0.0172	1720	1.6735	-
0.0173	1730	1.6442	-
0.0174	1740	1.6627	-
0.0175	1750	1.6752	-
0.0176	1760	1.6622	-
0.0177	1770	1.6472	-
0.0178	1780	1.6693	-
0.0179	1790	1.6684	-
0.018	1800	1.6578	0.5684
0.0181	1810	1.6298	-
0.0182	1820	1.6254	-
0.0183	1830	1.6306	-
0.0184	1840	1.6541	-
0.0185	1850	1.6412	-
0.0186	1860	1.6555	-
0.0187	1870	1.645	-
0.0188	1880	1.6459	-
0.0189	1890	1.6397	-
0.019	1900	1.6616	0.5806
0.0191	1910	1.6332	-
0.0192	1920	1.6484	-
0.0193	1930	1.6369	-
0.0194	1940	1.6319	-
0.0195	1950	1.637	-
0.0196	1960	1.6462	-
0.0197	1970	1.6294	-
0.0198	1980	1.6569	-
0.0199	1990	1.6217	-
0.02	2000	1.6364	0.5716
0.0201	2010	1.6423	-
0.0202	2020	1.6516	-
0.0203	2030	1.6221	-
0.0204	2040	1.6516	-
0.0205	2050	1.6496	-
0.0206	2060	1.6287	-
0.0207	2070	1.6076	-
0.0208	2080	1.6422	-
0.0209	2090	1.5888	-
0.021	2100	1.6276	0.5798
0.0211	2110	1.6335	-
0.0212	2120	1.641	-
0.0213	2130	1.6342	-
0.0214	2140	1.6394	-
0.0215	2150	1.628	-
0.0216	2160	1.6099	-
0.0217	2170	1.6256	-
0.0218	2180	1.6165	-
0.0219	2190	1.6209	-
0.022	2200	1.624	0.5790
0.0221	2210	1.6192	-
0.0222	2220	1.6196	-
0.0223	2230	1.6199	-
0.0224	2240	1.6233	-
0.0225	2250	1.6083	-
0.0226	2260	1.6282	-
0.0227	2270	1.6336	-
0.0228	2280	1.6221	-
0.0229	2290	1.6216	-
0.023	2300	1.5997	0.5803
0.0231	2310	1.6048	-
0.0232	2320	1.6129	-
0.0233	2330	1.616	-
0.0234	2340	1.6106	-
0.0235	2350	1.6201	-
0.0236	2360	1.6092	-
0.0237	2370	1.6109	-
0.0238	2380	1.6062	-
0.0239	2390	1.622	-
0.024	2400	1.616	0.5737
0.0241	2410	1.6261	-
0.0242	2420	1.5979	-
0.0243	2430	1.5988	-
0.0244	2440	1.5813	-
0.0245	2450	1.6147	-
0.0246	2460	1.6204	-
0.0247	2470	1.5962	-
0.0248	2480	1.6243	-
0.0249	2490	1.6151	-
0.025	2500	1.6171	0.5817
0.0251	2510	1.5873	-
0.0252	2520	1.5772	-
0.0253	2530	1.6069	-
0.0254	2540	1.6086	-
0.0255	2550	1.6028	-
0.0256	2560	1.6031	-
0.0257	2570	1.5907	-
0.0258	2580	1.6073	-
0.0259	2590	1.6105	-
0.026	2600	1.608	0.5771
0.0261	2610	1.6048	-
0.0262	2620	1.6119	-
0.0263	2630	1.5895	-
0.0264	2640	1.6335	-
0.0265	2650	1.5815	-
0.0266	2660	1.6071	-
0.0267	2670	1.6094	-
0.0268	2680	1.6241	-
0.0269	2690	1.6222	-
0.027	2700	1.597	0.5758
0.0271	2710	1.6102	-
0.0272	2720	1.582	-
0.0273	2730	1.5816	-
0.0274	2740	1.5896	-
0.0275	2750	1.5892	-
0.0276	2760	1.5979	-
0.0277	2770	1.6292	-
0.0278	2780	1.5712	-
0.0279	2790	1.5879	-
0.028	2800	1.6054	0.5767
0.0281	2810	1.5794	-
0.0282	2820	1.5901	-
0.0283	2830	1.5991	-
0.0284	2840	1.5806	-
0.0285	2850	1.6019	-
0.0286	2860	1.6132	-
0.0287	2870	1.5989	-
0.0288	2880	1.5878	-
0.0289	2890	1.5863	-
0.029	2900	1.6141	0.5810
0.0291	2910	1.5756	-
0.0292	2920	1.5885	-
0.0293	2930	1.5756	-
0.0294	2940	1.5793	-
0.0295	2950	1.5685	-
0.0296	2960	1.5996	-
0.0297	2970	1.5893	-
0.0298	2980	1.5791	-
0.0299	2990	1.5842	-
0.03	3000	1.5892	0.5810
0.0301	3010	1.5705	-
0.0302	3020	1.5865	-
0.0303	3030	1.5849	-
0.0304	3040	1.5873	-
0.0305	3050	1.5921	-
0.0306	3060	1.5991	-
0.0307	3070	1.5886	-
0.0308	3080	1.5752	-
0.0309	3090	1.5846	-
0.031	3100	1.5873	0.5766
0.0311	3110	1.5704	-
0.0312	3120	1.5963	-
0.0313	3130	1.5829	-
0.0314	3140	1.6	-
0.0315	3150	1.5741	-
0.0316	3160	1.5749	-
0.0317	3170	1.5736	-
0.0318	3180	1.5531	-
0.0319	3190	1.5881	-
0.032	3200	1.5873	0.5802
0.0321	3210	1.5842	-
0.0322	3220	1.5745	-
0.0323	3230	1.5668	-
0.0324	3240	1.5633	-
0.0325	3250	1.5696	-
0.0326	3260	1.5627	-
0.0327	3270	1.5761	-
0.0328	3280	1.5736	-
0.0329	3290	1.5668	-
0.033	3300	1.5575	0.5774
0.0331	3310	1.5599	-
0.0332	3320	1.5706	-
0.0333	3330	1.5845	-
0.0334	3340	1.5715	-
0.0335	3350	1.5809	-
0.0336	3360	1.5526	-
0.0337	3370	1.5568	-
0.0338	3380	1.5902	-
0.0339	3390	1.5789	-
0.034	3400	1.5848	0.5745
0.0341	3410	1.5772	-
0.0342	3420	1.5785	-
0.0343	3430	1.562	-
0.0344	3440	1.571	-
0.0345	3450	1.5546	-
0.0346	3460	1.5571	-
0.0347	3470	1.5519	-
0.0348	3480	1.5436	-
0.0349	3490	1.5607	-
0.035	3500	1.5546	0.5875
0.0351	3510	1.5716	-
0.0352	3520	1.5693	-
0.0353	3530	1.5597	-
0.0354	3540	1.599	-
0.0355	3550	1.5628	-
0.0356	3560	1.5615	-
0.0357	3570	1.5648	-
0.0358	3580	1.5563	-
0.0359	3590	1.5568	-
0.036	3600	1.5624	0.5863
0.0361	3610	1.5521	-
0.0362	3620	1.5528	-
0.0363	3630	1.5726	-
0.0364	3640	1.5666	-
0.0365	3650	1.5435	-
0.0366	3660	1.5654	-
0.0367	3670	1.5702	-
0.0368	3680	1.5627	-
0.0369	3690	1.5513	-
0.037	3700	1.5489	0.5826
0.0371	3710	1.5779	-
0.0372	3720	1.5654	-
0.0373	3730	1.5667	-
0.0374	3740	1.5577	-
0.0375	3750	1.5569	-
0.0376	3760	1.5466	-
0.0377	3770	1.5607	-
0.0378	3780	1.5619	-
0.0379	3790	1.5558	-
0.038	3800	1.5461	0.5805
0.0381	3810	1.5294	-
0.0382	3820	1.5604	-
0.0383	3830	1.559	-
0.0384	3840	1.5574	-
0.0385	3850	1.5621	-
0.0386	3860	1.5592	-
0.0387	3870	1.5425	-
0.0388	3880	1.5746	-
0.0389	3890	1.5508	-
0.039	3900	1.5563	0.5808
0.0391	3910	1.5515	-
0.0392	3920	1.5714	-
0.0393	3930	1.5479	-
0.0394	3940	1.556	-
0.0395	3950	1.5357	-
0.0396	3960	1.5375	-
0.0397	3970	1.5596	-
0.0398	3980	1.5426	-
0.0399	3990	1.553	-
0.04	4000	1.5766	0.5803
0.0401	4010	1.571	-
0.0402	4020	1.5331	-
0.0403	4030	1.5539	-
0.0404	4040	1.5608	-
0.0405	4050	1.5583	-
0.0406	4060	1.5277	-
0.0407	4070	1.5358	-
0.0408	4080	1.5524	-
0.0409	4090	1.559	-
0.041	4100	1.5683	0.5860
0.0411	4110	1.5497	-
0.0412	4120	1.5762	-
0.0413	4130	1.5585	-
0.0414	4140	1.5505	-
0.0415	4150	1.5411	-
0.0416	4160	1.5459	-
0.0417	4170	1.5625	-
0.0418	4180	1.5425	-
0.0419	4190	1.5554	-
0.042	4200	1.5193	0.5806
0.0421	4210	1.5098	-
0.0422	4220	1.5474	-
0.0423	4230	1.5352	-
0.0424	4240	1.5536	-
0.0425	4250	1.5549	-
0.0426	4260	1.5207	-
0.0427	4270	1.5641	-
0.0428	4280	1.5398	-
0.0429	4290	1.5419	-
0.043	4300	1.5311	0.5869
0.0431	4310	1.5437	-
0.0432	4320	1.5334	-
0.0433	4330	1.5271	-
0.0434	4340	1.5199	-
0.0435	4350	1.5526	-
0.0436	4360	1.5478	-
0.0437	4370	1.518	-
0.0438	4380	1.5441	-
0.0439	4390	1.5341	-
0.044	4400	1.542	0.5851
0.0441	4410	1.5381	-
0.0442	4420	1.5142	-
0.0443	4430	1.5514	-
0.0444	4440	1.5396	-
0.0445	4450	1.526	-
0.0446	4460	1.5374	-
0.0447	4470	1.5231	-
0.0448	4480	1.53	-
0.0449	4490	1.5281	-
0.045	4500	1.526	0.5844
0.0451	4510	1.5333	-
0.0452	4520	1.5664	-
0.0453	4530	1.5342	-
0.0454	4540	1.5438	-
0.0455	4550	1.5302	-
0.0456	4560	1.5299	-
0.0457	4570	1.5402	-
0.0458	4580	1.5455	-
0.0459	4590	1.5323	-
0.046	4600	1.5086	0.5748
0.0461	4610	1.5361	-
0.0462	4620	1.5181	-
0.0463	4630	1.5567	-
0.0464	4640	1.5329	-
0.0465	4650	1.5218	-
0.0466	4660	1.5386	-
0.0467	4670	1.5299	-
0.0468	4680	1.5175	-
0.0469	4690	1.5482	-
0.047	4700	1.5186	0.5768
0.0471	4710	1.5537	-
0.0472	4720	1.5023	-
0.0473	4730	1.546	-
0.0474	4740	1.5342	-
0.0475	4750	1.5237	-
0.0476	4760	1.5347	-
0.0477	4770	1.5358	-
0.0478	4780	1.5421	-
0.0479	4790	1.5263	-
0.048	4800	1.5227	0.5840
0.0481	4810	1.5238	-
0.0482	4820	1.5106	-
0.0483	4830	1.5257	-
0.0484	4840	1.5272	-
0.0485	4850	1.5218	-
0.0486	4860	1.5156	-
0.0487	4870	1.5325	-
0.0488	4880	1.5442	-
0.0489	4890	1.5442	-
0.049	4900	1.533	0.5726
0.0491	4910	1.5252	-
0.0492	4920	1.5163	-
0.0493	4930	1.5229	-
0.0494	4940	1.5241	-
0.0495	4950	1.5177	-
0.0496	4960	1.5359	-
0.0497	4970	1.5362	-
0.0498	4980	1.5192	-
0.0499	4990	1.518	-
0.05	5000	1.5283	0.5862
0.0501	5010	1.5571	-
0.0502	5020	1.5389	-
0.0503	5030	1.5133	-
0.0504	5040	1.5158	-
0.0505	5050	1.5297	-
0.0506	5060	1.5509	-
0.0507	5070	1.501	-
0.0508	5080	1.5143	-
0.0509	5090	1.5308	-
0.051	5100	1.5395	0.5756
0.0511	5110	1.5036	-
0.0512	5120	1.5277	-
0.0513	5130	1.5061	-
0.0514	5140	1.5062	-
0.0515	5150	1.504	-
0.0516	5160	1.5281	-
0.0517	5170	1.5255	-
0.0518	5180	1.5	-
0.0519	5190	1.5259	-
0.052	5200	1.527	0.5816
0.0521	5210	1.5186	-
0.0522	5220	1.5154	-
0.0523	5230	1.5195	-
0.0524	5240	1.5075	-
0.0525	5250	1.5263	-
0.0526	5260	1.5234	-
0.0527	5270	1.5135	-
0.0528	5280	1.5358	-
0.0529	5290	1.5168	-
0.053	5300	1.5179	0.5807
0.0531	5310	1.5233	-
0.0532	5320	1.5481	-
0.0533	5330	1.5268	-
0.0534	5340	1.5271	-
0.0535	5350	1.4983	-
0.0536	5360	1.5225	-
0.0537	5370	1.53	-
0.0538	5380	1.5065	-
0.0539	5390	1.5263	-
0.054	5400	1.5085	0.5796
0.0541	5410	1.5305	-
0.0542	5420	1.5397	-
0.0543	5430	1.5234	-
0.0544	5440	1.552	-
0.0545	5450	1.5103	-
0.0546	5460	1.5328	-
0.0547	5470	1.5057	-
0.0548	5480	1.5222	-
0.0549	5490	1.5196	-
0.055	5500	1.5166	0.5809
0.0551	5510	1.5084	-
0.0552	5520	1.511	-
0.0553	5530	1.5288	-
0.0554	5540	1.514	-
0.0555	5550	1.5085	-
0.0556	5560	1.5317	-
0.0557	5570	1.5045	-
0.0558	5580	1.5029	-
0.0559	5590	1.5074	-
0.056	5600	1.5273	0.5897
0.0561	5610	1.5231	-
0.0562	5620	1.5076	-
0.0563	5630	1.5214	-
0.0564	5640	1.494	-
0.0565	5650	1.4888	-
0.0566	5660	1.5198	-
0.0567	5670	1.5032	-
0.0568	5680	1.5044	-
0.0569	5690	1.5353	-
0.057	5700	1.5278	0.5819
0.0571	5710	1.5271	-
0.0572	5720	1.5428	-
0.0573	5730	1.5173	-
0.0574	5740	1.5014	-
0.0575	5750	1.5198	-
0.0576	5760	1.5086	-
0.0577	5770	1.543	-
0.0578	5780	1.503	-
0.0579	5790	1.5198	-
0.058	5800	1.4969	0.5786
0.0581	5810	1.5054	-
0.0582	5820	1.5446	-
0.0583	5830	1.5164	-
0.0584	5840	1.5278	-
0.0585	5850	1.4902	-
0.0586	5860	1.5271	-
0.0587	5870	1.5026	-
0.0588	5880	1.5284	-
0.0589	5890	1.5023	-
0.059	5900	1.5138	0.5839
0.0591	5910	1.5136	-
0.0592	5920	1.5308	-
0.0593	5930	1.4936	-
0.0594	5940	1.5274	-
0.0595	5950	1.4917	-
0.0596	5960	1.5362	-
0.0597	5970	1.5135	-
0.0598	5980	1.5217	-
0.0599	5990	1.5234	-
0.06	6000	1.5168	0.5768
0.0601	6010	1.5203	-
0.0602	6020	1.5263	-
0.0603	6030	1.5236	-
0.0604	6040	1.5128	-
0.0605	6050	1.521	-
0.0606	6060	1.515	-
0.0607	6070	1.5111	-
0.0608	6080	1.523	-
0.0609	6090	1.4804	-
0.061	6100	1.5065	0.5790

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}
}