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_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, 0.8875, 0.8498],
#         [0.8875, 1.0000, 0.7767],
#         [0.8498, 0.7767, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: validation_retrieval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.48
cosine_accuracy@3	0.75
cosine_accuracy@5	0.83
cosine_accuracy@10	0.91
cosine_accuracy@50	1.0
cosine_accuracy@100	1.0
cosine_accuracy@300	1.0
cosine_accuracy@500	1.0
cosine_accuracy@1000	1.0
cosine_precision@1	0.48
cosine_precision@3	0.4033
cosine_precision@5	0.35
cosine_precision@10	0.267
cosine_precision@50	0.118
cosine_precision@100	0.0735
cosine_precision@300	0.0319
cosine_precision@500	0.0209
cosine_precision@1000	0.0116
cosine_recall@1	0.1045
cosine_recall@3	0.2452
cosine_recall@5	0.3285
cosine_recall@10	0.3966
cosine_recall@50	0.6415
cosine_recall@100	0.713
cosine_recall@300	0.8112
cosine_recall@500	0.8641
cosine_recall@1000	0.9148
cosine_ndcg@1	0.48
cosine_ndcg@3	0.4653
cosine_ndcg@5	0.4652
cosine_ndcg@10	0.4503
cosine_ndcg@50	0.5001
cosine_ndcg@100	0.5276
cosine_ndcg@300	0.5644
cosine_ndcg@500	0.5794
cosine_ndcg@1000	0.5932
cosine_mrr@1	0.48
cosine_mrr@3	0.6
cosine_mrr@5	0.6185
cosine_mrr@10	0.6311
cosine_mrr@50	0.6363
cosine_mrr@100	0.6363
cosine_mrr@300	0.6363
cosine_mrr@500	0.6363
cosine_mrr@1000	0.6363
cosine_map@1	0.48
cosine_map@3	0.3914
cosine_map@5	0.3689
cosine_map@10	0.3293
cosine_map@50	0.3227
cosine_map@100	0.3316
cosine_map@300	0.3401
cosine_map@500	0.3418
cosine_map@1000	0.3431

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@1000
1e-05	1	5.2968	-
0.0001	10	3.8535	-
0.0002	20	2.8728	-
0.0003	30	2.5666	-
0.0004	40	2.4087	-
0.0005	50	2.3218	-
0.0006	60	2.3077	-
0.0007	70	2.2544	-
0.0008	80	2.2092	-
0.0009	90	2.1864	-
0.001	100	2.1891	0.5524
0.0011	110	2.147	-
0.0012	120	2.1584	-
0.0013	130	2.1306	-
0.0014	140	2.1056	-
0.0015	150	2.0799	-
0.0016	160	2.0855	-
0.0017	170	2.0765	-
0.0018	180	2.0571	-
0.0019	190	2.0685	-
0.002	200	2.0674	0.5699
0.0021	210	2.0417	-
0.0022	220	2.0162	-
0.0023	230	2.0255	-
0.0024	240	2.004	-
0.0025	250	2.0104	-
0.0026	260	2.0237	-
0.0027	270	2.024	-
0.0028	280	1.9991	-
0.0029	290	1.9902	-
0.003	300	1.9874	0.5720
0.0031	310	1.9881	-
0.0032	320	1.9763	-
0.0033	330	1.956	-
0.0034	340	1.9452	-
0.0035	350	1.9528	-
0.0036	360	1.942	-
0.0037	370	1.9366	-
0.0038	380	1.9353	-
0.0039	390	1.9187	-
0.004	400	1.9261	0.5775
0.0041	410	1.9484	-
0.0042	420	1.9329	-
0.0043	430	1.905	-
0.0044	440	1.8921	-
0.0045	450	1.9125	-
0.0046	460	1.9225	-
0.0047	470	1.9342	-
0.0048	480	1.9407	-
0.0049	490	1.8906	-
0.005	500	1.8838	0.5749
0.0051	510	1.8975	-
0.0052	520	1.8947	-
0.0053	530	1.8862	-
0.0054	540	1.883	-
0.0055	550	1.887	-
0.0056	560	1.8744	-
0.0057	570	1.8847	-
0.0058	580	1.8845	-
0.0059	590	1.8394	-
0.006	600	1.8393	0.5778
0.0061	610	1.8681	-
0.0062	620	1.861	-
0.0063	630	1.8681	-
0.0064	640	1.8426	-
0.0065	650	1.8529	-
0.0066	660	1.8484	-
0.0067	670	1.8734	-
0.0068	680	1.8558	-
0.0069	690	1.8561	-
0.007	700	1.8593	0.5770
0.0071	710	1.8802	-
0.0072	720	1.8236	-
0.0073	730	1.8151	-
0.0074	740	1.8519	-
0.0075	750	1.8278	-
0.0076	760	1.831	-
0.0077	770	1.839	-
0.0078	780	1.8381	-
0.0079	790	1.8083	-
0.008	800	1.8583	0.5727
0.0081	810	1.7964	-
0.0082	820	1.8388	-
0.0083	830	1.8153	-
0.0084	840	1.8093	-
0.0085	850	1.8117	-
0.0086	860	1.8195	-
0.0087	870	1.8004	-
0.0088	880	1.8161	-
0.0089	890	1.8243	-
0.009	900	1.8175	0.5789
0.0091	910	1.7898	-
0.0092	920	1.8152	-
0.0093	930	1.7938	-
0.0094	940	1.806	-
0.0095	950	1.7863	-
0.0096	960	1.806	-
0.0097	970	1.8174	-
0.0098	980	1.7857	-
0.0099	990	1.7762	-
0.01	1000	1.7772	0.5770
0.0101	1010	1.8122	-
0.0102	1020	1.7781	-
0.0103	1030	1.7767	-
0.0104	1040	1.7838	-
0.0105	1050	1.7957	-
0.0106	1060	1.7863	-
0.0107	1070	1.7836	-
0.0108	1080	1.8069	-
0.0109	1090	1.7895	-
0.011	1100	1.7675	0.5738
0.0111	1110	1.7903	-
0.0112	1120	1.7693	-
0.0113	1130	1.7936	-
0.0114	1140	1.7837	-
0.0115	1150	1.7602	-
0.0116	1160	1.7585	-
0.0117	1170	1.786	-
0.0118	1180	1.7642	-
0.0119	1190	1.7492	-
0.012	1200	1.7886	0.5819
0.0121	1210	1.7821	-
0.0122	1220	1.7665	-
0.0123	1230	1.7616	-
0.0124	1240	1.7566	-
0.0125	1250	1.7537	-
0.0126	1260	1.752	-
0.0127	1270	1.7552	-
0.0128	1280	1.7816	-
0.0129	1290	1.7658	-
0.013	1300	1.757	0.5780
0.0131	1310	1.7682	-
0.0132	1320	1.749	-
0.0133	1330	1.756	-
0.0134	1340	1.7569	-
0.0135	1350	1.7662	-
0.0136	1360	1.7657	-
0.0137	1370	1.76	-
0.0138	1380	1.7746	-
0.0139	1390	1.7438	-
0.014	1400	1.7309	0.5842
0.0141	1410	1.737	-
0.0142	1420	1.7566	-
0.0143	1430	1.7443	-
0.0144	1440	1.7256	-
0.0145	1450	1.7396	-
0.0146	1460	1.7326	-
0.0147	1470	1.7497	-
0.0148	1480	1.7187	-
0.0149	1490	1.7168	-
0.015	1500	1.7346	0.5754
0.0151	1510	1.7434	-
0.0152	1520	1.7454	-
0.0153	1530	1.7721	-
0.0154	1540	1.7264	-
0.0155	1550	1.7168	-
0.0156	1560	1.7341	-
0.0157	1570	1.7316	-
0.0158	1580	1.7282	-
0.0159	1590	1.7254	-
0.016	1600	1.7313	0.5756
0.0161	1610	1.7363	-
0.0162	1620	1.72	-
0.0163	1630	1.7149	-
0.0164	1640	1.7408	-
0.0165	1650	1.7162	-
0.0166	1660	1.7096	-
0.0167	1670	1.7476	-
0.0168	1680	1.7133	-
0.0169	1690	1.7207	-
0.017	1700	1.7016	0.5812
0.0171	1710	1.7274	-
0.0172	1720	1.7244	-
0.0173	1730	1.7153	-
0.0174	1740	1.724	-
0.0175	1750	1.6935	-
0.0176	1760	1.7157	-
0.0177	1770	1.7182	-
0.0178	1780	1.6985	-
0.0179	1790	1.728	-
0.018	1800	1.7048	0.5736
0.0181	1810	1.7174	-
0.0182	1820	1.7049	-
0.0183	1830	1.6976	-
0.0184	1840	1.6829	-
0.0185	1850	1.7058	-
0.0186	1860	1.7111	-
0.0187	1870	1.6971	-
0.0188	1880	1.6871	-
0.0189	1890	1.7321	-
0.019	1900	1.7013	0.5808
0.0191	1910	1.7195	-
0.0192	1920	1.712	-
0.0193	1930	1.7314	-
0.0194	1940	1.6948	-
0.0195	1950	1.7226	-
0.0196	1960	1.7175	-
0.0197	1970	1.6685	-
0.0198	1980	1.682	-
0.0199	1990	1.7187	-
0.02	2000	1.6877	0.5853
0.0201	2010	1.7166	-
0.0202	2020	1.6874	-
0.0203	2030	1.6903	-
0.0204	2040	1.6716	-
0.0205	2050	1.6645	-
0.0206	2060	1.7001	-
0.0207	2070	1.7103	-
0.0208	2080	1.6864	-
0.0209	2090	1.6829	-
0.021	2100	1.6797	0.5792
0.0211	2110	1.705	-
0.0212	2120	1.6723	-
0.0213	2130	1.7026	-
0.0214	2140	1.6944	-
0.0215	2150	1.6667	-
0.0216	2160	1.6885	-
0.0217	2170	1.6794	-
0.0218	2180	1.7	-
0.0219	2190	1.6811	-
0.022	2200	1.678	0.5776
0.0221	2210	1.6918	-
0.0222	2220	1.661	-
0.0223	2230	1.6649	-
0.0224	2240	1.6922	-
0.0225	2250	1.6883	-
0.0226	2260	1.6865	-
0.0227	2270	1.6667	-
0.0228	2280	1.6826	-
0.0229	2290	1.6757	-
0.023	2300	1.6721	0.5869
0.0231	2310	1.6804	-
0.0232	2320	1.6953	-
0.0233	2330	1.6749	-
0.0234	2340	1.6995	-
0.0235	2350	1.6537	-
0.0236	2360	1.665	-
0.0237	2370	1.6573	-
0.0238	2380	1.6817	-
0.0239	2390	1.6993	-
0.024	2400	1.6729	0.5865
0.0241	2410	1.661	-
0.0242	2420	1.6961	-
0.0243	2430	1.6854	-
0.0244	2440	1.6338	-
0.0245	2450	1.6674	-
0.0246	2460	1.688	-
0.0247	2470	1.6557	-
0.0248	2480	1.6759	-
0.0249	2490	1.6663	-
0.025	2500	1.6728	0.5827
0.0251	2510	1.6914	-
0.0252	2520	1.6699	-
0.0253	2530	1.6548	-
0.0254	2540	1.6506	-
0.0255	2550	1.6598	-
0.0256	2560	1.6581	-
0.0257	2570	1.6747	-
0.0258	2580	1.6735	-
0.0259	2590	1.6723	-
0.026	2600	1.6458	0.5821
0.0261	2610	1.6573	-
0.0262	2620	1.6641	-
0.0263	2630	1.6652	-
0.0264	2640	1.6603	-
0.0265	2650	1.6492	-
0.0266	2660	1.6623	-
0.0267	2670	1.6192	-
0.0268	2680	1.6466	-
0.0269	2690	1.6413	-
0.027	2700	1.65	0.5831
0.0271	2710	1.6513	-
0.0272	2720	1.648	-
0.0273	2730	1.6601	-
0.0274	2740	1.639	-
0.0275	2750	1.6674	-
0.0276	2760	1.6653	-
0.0277	2770	1.6436	-
0.0278	2780	1.6546	-
0.0279	2790	1.6717	-
0.028	2800	1.6599	0.5880
0.0281	2810	1.6431	-
0.0282	2820	1.6496	-
0.0283	2830	1.6508	-
0.0284	2840	1.6652	-
0.0285	2850	1.6422	-
0.0286	2860	1.6524	-
0.0287	2870	1.6645	-
0.0288	2880	1.6197	-
0.0289	2890	1.6287	-
0.029	2900	1.6306	0.5828
0.0291	2910	1.6345	-
0.0292	2920	1.6377	-
0.0293	2930	1.6331	-
0.0294	2940	1.6509	-
0.0295	2950	1.6376	-
0.0296	2960	1.6565	-
0.0297	2970	1.6348	-
0.0298	2980	1.6289	-
0.0299	2990	1.6513	-
0.03	3000	1.6369	0.5843
0.0301	3010	1.6571	-
0.0302	3020	1.63	-
0.0303	3030	1.6581	-
0.0304	3040	1.6373	-
0.0305	3050	1.6178	-
0.0306	3060	1.6278	-
0.0307	3070	1.6588	-
0.0308	3080	1.6284	-
0.0309	3090	1.6318	-
0.031	3100	1.6184	0.5828
0.0311	3110	1.6406	-
0.0312	3120	1.6441	-
0.0313	3130	1.6367	-
0.0314	3140	1.6447	-
0.0315	3150	1.6465	-
0.0316	3160	1.6482	-
0.0317	3170	1.631	-
0.0318	3180	1.6315	-
0.0319	3190	1.6461	-
0.032	3200	1.6255	0.5881
0.0321	3210	1.6182	-
0.0322	3220	1.6314	-
0.0323	3230	1.6455	-
0.0324	3240	1.631	-
0.0325	3250	1.6177	-
0.0326	3260	1.6189	-
0.0327	3270	1.6439	-
0.0328	3280	1.6109	-
0.0329	3290	1.636	-
0.033	3300	1.6238	0.5853
0.0331	3310	1.6129	-
0.0332	3320	1.6236	-
0.0333	3330	1.6399	-
0.0334	3340	1.6145	-
0.0335	3350	1.6038	-
0.0336	3360	1.6332	-
0.0337	3370	1.6351	-
0.0338	3380	1.6107	-
0.0339	3390	1.6216	-
0.034	3400	1.6039	0.5845
0.0341	3410	1.6425	-
0.0342	3420	1.6381	-
0.0343	3430	1.6282	-
0.0344	3440	1.6055	-
0.0345	3450	1.6216	-
0.0346	3460	1.5988	-
0.0347	3470	1.6198	-
0.0348	3480	1.6161	-
0.0349	3490	1.6173	-
0.035	3500	1.6266	0.5814
0.0351	3510	1.6281	-
0.0352	3520	1.6161	-
0.0353	3530	1.6075	-
0.0354	3540	1.6345	-
0.0355	3550	1.6118	-
0.0356	3560	1.6129	-
0.0357	3570	1.5996	-
0.0358	3580	1.606	-
0.0359	3590	1.5885	-
0.036	3600	1.6284	0.5780
0.0361	3610	1.5932	-
0.0362	3620	1.609	-
0.0363	3630	1.6114	-
0.0364	3640	1.6115	-
0.0365	3650	1.6414	-
0.0366	3660	1.6153	-
0.0367	3670	1.6069	-
0.0368	3680	1.6318	-
0.0369	3690	1.6296	-
0.037	3700	1.6169	0.5787
0.0371	3710	1.6131	-
0.0372	3720	1.6193	-
0.0373	3730	1.6169	-
0.0374	3740	1.5862	-
0.0375	3750	1.6023	-
0.0376	3760	1.5968	-
0.0377	3770	1.6045	-
0.0378	3780	1.6055	-
0.0379	3790	1.6134	-
0.038	3800	1.6252	0.5872
0.0381	3810	1.6317	-
0.0382	3820	1.6354	-
0.0383	3830	1.6271	-
0.0384	3840	1.6148	-
0.0385	3850	1.5861	-
0.0386	3860	1.5795	-
0.0387	3870	1.5897	-
0.0388	3880	1.6318	-
0.0389	3890	1.5974	-
0.039	3900	1.6191	0.5984
0.0391	3910	1.5942	-
0.0392	3920	1.5957	-
0.0393	3930	1.6117	-
0.0394	3940	1.587	-
0.0395	3950	1.5731	-
0.0396	3960	1.5859	-
0.0397	3970	1.5989	-
0.0398	3980	1.6053	-
0.0399	3990	1.6049	-
0.04	4000	1.5841	0.5951
0.0401	4010	1.5917	-
0.0402	4020	1.5812	-
0.0403	4030	1.5925	-
0.0404	4040	1.5946	-
0.0405	4050	1.624	-
0.0406	4060	1.6196	-
0.0407	4070	1.6108	-
0.0408	4080	1.6052	-
0.0409	4090	1.6071	-
0.041	4100	1.6079	0.5856
0.0411	4110	1.585	-
0.0412	4120	1.6135	-
0.0413	4130	1.5851	-
0.0414	4140	1.571	-
0.0415	4150	1.6236	-
0.0416	4160	1.5602	-
0.0417	4170	1.5728	-
0.0418	4180	1.6029	-
0.0419	4190	1.6013	-
0.042	4200	1.5996	0.5878
0.0421	4210	1.5919	-
0.0422	4220	1.6011	-
0.0423	4230	1.5956	-
0.0424	4240	1.5653	-
0.0425	4250	1.5781	-
0.0426	4260	1.6026	-
0.0427	4270	1.6052	-
0.0428	4280	1.574	-
0.0429	4290	1.5673	-
0.043	4300	1.6119	0.5887
0.0431	4310	1.5932	-
0.0432	4320	1.5869	-
0.0433	4330	1.5837	-
0.0434	4340	1.5919	-
0.0435	4350	1.6081	-
0.0436	4360	1.5961	-
0.0437	4370	1.6008	-
0.0438	4380	1.5922	-
0.0439	4390	1.5869	-
0.044	4400	1.5747	0.5823
0.0441	4410	1.6002	-
0.0442	4420	1.589	-
0.0443	4430	1.5874	-
0.0444	4440	1.5857	-
0.0445	4450	1.6128	-
0.0446	4460	1.5746	-
0.0447	4470	1.5886	-
0.0448	4480	1.5857	-
0.0449	4490	1.5773	-
0.045	4500	1.5894	0.5848
0.0451	4510	1.5783	-
0.0452	4520	1.5973	-
0.0453	4530	1.5958	-
0.0454	4540	1.6042	-
0.0455	4550	1.5834	-
0.0456	4560	1.5892	-
0.0457	4570	1.5671	-
0.0458	4580	1.6019	-
0.0459	4590	1.5918	-
0.046	4600	1.5903	0.5942
0.0461	4610	1.5806	-
0.0462	4620	1.588	-
0.0463	4630	1.5829	-
0.0464	4640	1.5928	-
0.0465	4650	1.5891	-
0.0466	4660	1.5676	-
0.0467	4670	1.589	-
0.0468	4680	1.5614	-
0.0469	4690	1.5832	-
0.047	4700	1.579	0.5932
0.0471	4710	1.5846	-
0.0472	4720	1.5606	-
0.0473	4730	1.5911	-
0.0474	4740	1.5905	-
0.0475	4750	1.5924	-
0.0476	4760	1.5907	-
0.0477	4770	1.5964	-
0.0478	4780	1.5839	-
0.0479	4790	1.5726	-
0.048	4800	1.5751	0.5886
0.0481	4810	1.588	-
0.0482	4820	1.5634	-
0.0483	4830	1.5817	-
0.0484	4840	1.5722	-
0.0485	4850	1.5734	-
0.0486	4860	1.5849	-
0.0487	4870	1.5437	-
0.0488	4880	1.5691	-
0.0489	4890	1.5677	-
0.049	4900	1.5762	0.5893
0.0491	4910	1.5952	-
0.0492	4920	1.5928	-
0.0493	4930	1.5647	-
0.0494	4940	1.5745	-
0.0495	4950	1.563	-
0.0496	4960	1.5937	-
0.0497	4970	1.5836	-
0.0498	4980	1.5937	-
0.0499	4990	1.5767	-
0.05	5000	1.5851	0.5878
0.0501	5010	1.5541	-
0.0502	5020	1.5694	-
0.0503	5030	1.5694	-
0.0504	5040	1.5831	-
0.0505	5050	1.5936	-
0.0506	5060	1.5883	-
0.0507	5070	1.5744	-
0.0508	5080	1.5887	-
0.0509	5090	1.5747	-
0.051	5100	1.5747	0.5850
0.0511	5110	1.5657	-
0.0512	5120	1.5678	-
0.0513	5130	1.5772	-
0.0514	5140	1.5752	-
0.0515	5150	1.5789	-
0.0516	5160	1.5753	-
0.0517	5170	1.5602	-
0.0518	5180	1.5564	-
0.0519	5190	1.5542	-
0.052	5200	1.5692	0.5968
0.0521	5210	1.5804	-
0.0522	5220	1.5891	-
0.0523	5230	1.5576	-
0.0524	5240	1.5932	-
0.0525	5250	1.5526	-
0.0526	5260	1.5747	-
0.0527	5270	1.5726	-
0.0528	5280	1.5681	-
0.0529	5290	1.5789	-
0.053	5300	1.5658	0.5962
0.0531	5310	1.5616	-
0.0532	5320	1.5406	-
0.0533	5330	1.5802	-
0.0534	5340	1.5707	-
0.0535	5350	1.538	-
0.0536	5360	1.5594	-
0.0537	5370	1.5607	-
0.0538	5380	1.559	-
0.0539	5390	1.5473	-
0.054	5400	1.5898	0.5947
0.0541	5410	1.5905	-
0.0542	5420	1.5371	-
0.0543	5430	1.5406	-
0.0544	5440	1.5802	-
0.0545	5450	1.5554	-
0.0546	5460	1.5514	-
0.0547	5470	1.5428	-
0.0548	5480	1.5781	-
0.0549	5490	1.5455	-
0.055	5500	1.5743	0.5955
0.0551	5510	1.5687	-
0.0552	5520	1.5412	-
0.0553	5530	1.5577	-
0.0554	5540	1.5674	-
0.0555	5550	1.5724	-
0.0556	5560	1.5483	-
0.0557	5570	1.5585	-
0.0558	5580	1.5687	-
0.0559	5590	1.5533	-
0.056	5600	1.5747	0.5890
0.0561	5610	1.5561	-
0.0562	5620	1.5457	-
0.0563	5630	1.5553	-
0.0564	5640	1.5714	-
0.0565	5650	1.5562	-
0.0566	5660	1.5809	-
0.0567	5670	1.5486	-
0.0568	5680	1.5492	-
0.0569	5690	1.5563	-
0.057	5700	1.5761	0.5924
0.0571	5710	1.538	-
0.0572	5720	1.5642	-
0.0573	5730	1.5448	-
0.0574	5740	1.5602	-
0.0575	5750	1.577	-
0.0576	5760	1.5605	-
0.0577	5770	1.5577	-
0.0578	5780	1.5415	-
0.0579	5790	1.5601	-
0.058	5800	1.5592	0.5979
0.0581	5810	1.5853	-
0.0582	5820	1.5558	-
0.0583	5830	1.577	-
0.0584	5840	1.5312	-
0.0585	5850	1.568	-
0.0586	5860	1.5672	-
0.0587	5870	1.5638	-
0.0588	5880	1.5582	-
0.0589	5890	1.5736	-
0.059	5900	1.5465	0.5951
0.0591	5910	1.549	-
0.0592	5920	1.5523	-
0.0593	5930	1.578	-
0.0594	5940	1.5481	-
0.0595	5950	1.5597	-
0.0596	5960	1.5495	-
0.0597	5970	1.5683	-
0.0598	5980	1.5389	-
0.0599	5990	1.5372	-
0.06	6000	1.5784	0.5918
0.0601	6010	1.5426	-
0.0602	6020	1.5614	-
0.0603	6030	1.5648	-
0.0604	6040	1.553	-
0.0605	6050	1.5679	-
0.0606	6060	1.5419	-
0.0607	6070	1.5632	-
0.0608	6080	1.547	-
0.0609	6090	1.5369	-
0.061	6100	1.5575	0.5851
0.0611	6110	1.5403	-
0.0612	6120	1.56	-
0.0613	6130	1.5423	-
0.0614	6140	1.5566	-
0.0615	6150	1.5647	-
0.0616	6160	1.5398	-
0.0617	6170	1.5862	-
0.0618	6180	1.5743	-
0.0619	6190	1.5491	-
0.062	6200	1.5596	0.5906
0.0621	6210	1.5695	-
0.0622	6220	1.5465	-
0.0623	6230	1.5806	-
0.0624	6240	1.5563	-
0.0625	6250	1.5661	-
0.0626	6260	1.5688	-
0.0627	6270	1.5609	-
0.0628	6280	1.5532	-
0.0629	6290	1.5342	-
0.063	6300	1.5729	0.5905
0.0631	6310	1.5388	-
0.0632	6320	1.5354	-
0.0633	6330	1.5475	-
0.0634	6340	1.5767	-
0.0635	6350	1.5179	-
0.0636	6360	1.5587	-
0.0637	6370	1.5475	-
0.0638	6380	1.5637	-
0.0639	6390	1.5457	-
0.064	6400	1.5585	0.5950
0.0641	6410	1.5626	-
0.0642	6420	1.559	-
0.0643	6430	1.5394	-
0.0644	6440	1.5433	-
0.0645	6450	1.5476	-
0.0646	6460	1.5472	-
0.0647	6470	1.5602	-
0.0648	6480	1.5576	-
0.0649	6490	1.5485	-
0.065	6500	1.572	0.5868
0.0651	6510	1.5382	-
0.0652	6520	1.5464	-
0.0653	6530	1.5374	-
0.0654	6540	1.5471	-
0.0655	6550	1.566	-
0.0656	6560	1.5504	-
0.0657	6570	1.5479	-
0.0658	6580	1.5479	-
0.0659	6590	1.5315	-
0.066	6600	1.5405	0.5951
0.0661	6610	1.5238	-
0.0662	6620	1.5464	-
0.0663	6630	1.5449	-
0.0664	6640	1.5689	-
0.0665	6650	1.5413	-
0.0666	6660	1.5295	-
1.0000	6670	1.5423	-
1.0001	6680	1.533	-
1.0002	6690	1.5597	-
1.0003	6700	1.5509	0.5862
1.0004	6710	1.5395	-
1.0005	6720	1.5548	-
1.0006	6730	1.5471	-
1.0007	6740	1.5418	-
1.0008	6750	1.5408	-
1.0009	6760	1.5325	-
1.0010	6770	1.562	-
1.0011	6780	1.531	-
1.0012	6790	1.5488	-
1.0013	6800	1.5353	0.5934
1.0014	6810	1.5139	-
1.0015	6820	1.5382	-
1.0016	6830	1.5313	-
1.0017	6840	1.5458	-
1.0018	6850	1.5222	-
1.0019	6860	1.5627	-
1.0020	6870	1.5425	-
1.0021	6880	1.5254	-
1.0022	6890	1.5246	-
1.0023	6900	1.5278	0.5969
1.0024	6910	1.5192	-
1.0025	6920	1.5449	-
1.0026	6930	1.5458	-
1.0027	6940	1.5419	-
1.0028	6950	1.5423	-
1.0029	6960	1.5214	-
1.0030	6970	1.542	-
1.0031	6980	1.5471	-
1.0032	6990	1.5391	-
1.0033	7000	1.5255	0.5900
1.0034	7010	1.5225	-
1.0035	7020	1.524	-
1.0036	7030	1.5172	-
1.0037	7040	1.5326	-
1.0038	7050	1.5184	-
1.0039	7060	1.5224	-
1.0040	7070	1.5454	-
1.0041	7080	1.5388	-
1.0042	7090	1.5299	-
1.0043	7100	1.5049	0.5911
1.0044	7110	1.5186	-
1.0045	7120	1.5244	-
1.0046	7130	1.5368	-
1.0047	7140	1.5644	-
1.0048	7150	1.5458	-
1.0049	7160	1.4983	-
1.0050	7170	1.5164	-
1.0051	7180	1.5336	-
1.0052	7190	1.5194	-
1.0053	7200	1.5239	0.5907
1.0054	7210	1.5272	-
1.0055	7220	1.5312	-
1.0056	7230	1.518	-
1.0057	7240	1.5372	-
1.0058	7250	1.5273	-
1.0059	7260	1.4783	-
1.0060	7270	1.5248	-
1.0061	7280	1.5309	-
1.0062	7290	1.5018	-
1.0063	7300	1.5297	0.5917
1.0064	7310	1.5049	-
1.0065	7320	1.5198	-
1.0066	7330	1.5292	-
1.0067	7340	1.534	-
1.0068	7350	1.5199	-
1.0069	7360	1.536	-
1.0070	7370	1.5387	-
1.0071	7380	1.5205	-
1.0072	7390	1.5103	-
1.0073	7400	1.5152	0.5982
1.0074	7410	1.5335	-
1.0075	7420	1.494	-
1.0076	7430	1.5371	-
1.0077	7440	1.5285	-
1.0078	7450	1.5127	-
1.0079	7460	1.5124	-
1.0080	7470	1.5451	-
1.0081	7480	1.4861	-
1.0082	7490	1.545	-
1.0083	7500	1.5099	0.5921
1.0084	7510	1.5218	-
1.0085	7520	1.5093	-
1.0086	7530	1.4986	-
1.0087	7540	1.5275	-
1.0088	7550	1.5214	-
1.0089	7560	1.5287	-
1.0090	7570	1.5177	-
1.0091	7580	1.5267	-
1.0092	7590	1.5075	-
1.0093	7600	1.498	0.5857
1.0094	7610	1.5222	-
1.0095	7620	1.5102	-
1.0096	7630	1.5235	-
1.0097	7640	1.5193	-
1.0098	7650	1.5142	-
1.0099	7660	1.4977	-
1.0100	7670	1.5076	-
1.0101	7680	1.5194	-
1.0102	7690	1.5137	-
1.0103	7700	1.4966	0.5862
1.0104	7710	1.53	-
1.0105	7720	1.5314	-
1.0106	7730	1.5065	-
1.0107	7740	1.5362	-
1.0108	7750	1.5387	-
1.0109	7760	1.5111	-
1.0110	7770	1.5197	-
1.0111	7780	1.5163	-
1.0112	7790	1.5261	-
1.0113	7800	1.5145	0.5940
1.0114	7810	1.5213	-
1.0115	7820	1.4953	-
1.0116	7830	1.5035	-
1.0117	7840	1.5324	-
1.0118	7850	1.4921	-
1.0119	7860	1.5148	-
1.0120	7870	1.5248	-
1.0121	7880	1.5356	-
1.0122	7890	1.5152	-
1.0123	7900	1.4958	0.5856
1.0124	7910	1.5051	-
1.0125	7920	1.5082	-
1.0126	7930	1.4854	-
1.0127	7940	1.5395	-
1.0128	7950	1.5422	-
1.0129	7960	1.5123	-
1.0130	7970	1.5043	-
1.0131	7980	1.529	-
1.0132	7990	1.512	-
1.0133	8000	1.5213	0.5875
1.0134	8010	1.5207	-
1.0135	8020	1.529	-
1.0136	8030	1.5129	-
1.0137	8040	1.5444	-
1.0138	8050	1.5217	-
1.0139	8060	1.5062	-
1.0140	8070	1.5027	-
1.0141	8080	1.4976	-
1.0142	8090	1.5288	-
1.0143	8100	1.5041	0.5896
1.0144	8110	1.512	-
1.0145	8120	1.5075	-
1.0146	8130	1.5106	-
1.0147	8140	1.4954	-
1.0148	8150	1.5073	-
1.0149	8160	1.4829	-
1.0150	8170	1.51	-
1.0151	8180	1.5124	-
1.0152	8190	1.5331	-
1.0153	8200	1.535	0.5902
1.0154	8210	1.4911	-
1.0155	8220	1.4932	-
1.0156	8230	1.499	-
1.0157	8240	1.5105	-
1.0158	8250	1.5103	-
1.0159	8260	1.5217	-
1.0160	8270	1.5192	-
1.0161	8280	1.5079	-
1.0162	8290	1.5121	-
1.0163	8300	1.5082	0.5925
1.0164	8310	1.5075	-
1.0165	8320	1.5011	-
1.0166	8330	1.5108	-
1.0167	8340	1.5279	-
1.0168	8350	1.4909	-
1.0169	8360	1.5049	-
1.0170	8370	1.5021	-
1.0171	8380	1.5166	-
1.0172	8390	1.5275	-
1.0173	8400	1.506	0.5906
1.0174	8410	1.5035	-
1.0175	8420	1.5007	-
1.0176	8430	1.5197	-
1.0177	8440	1.4934	-
1.0178	8450	1.5129	-
1.0179	8460	1.5338	-
1.0180	8470	1.4849	-
1.0181	8480	1.5154	-
1.0182	8490	1.5132	-
1.0183	8500	1.5038	0.5920
1.0184	8510	1.4842	-
1.0185	8520	1.5157	-
1.0186	8530	1.4997	-
1.0187	8540	1.5048	-
1.0188	8550	1.5073	-
1.0189	8560	1.5173	-
1.0190	8570	1.5163	-
1.0191	8580	1.5331	-
1.0192	8590	1.5194	-
1.0193	8600	1.518	0.5917
1.0194	8610	1.5177	-
1.0195	8620	1.5297	-
1.0196	8630	1.5094	-
1.0197	8640	1.4885	-
1.0198	8650	1.4925	-
1.0199	8660	1.5197	-
1.0200	8670	1.5027	-
1.0201	8680	1.519	-
1.0202	8690	1.4977	-
1.0203	8700	1.5008	0.5911
1.0204	8710	1.4869	-
1.0205	8720	1.4918	-
1.0206	8730	1.5238	-
1.0207	8740	1.5052	-
1.0208	8750	1.5011	-
1.0209	8760	1.5098	-
1.0210	8770	1.5069	-
1.0211	8780	1.513	-
1.0212	8790	1.5037	-
1.0213	8800	1.5138	0.5877
1.0214	8810	1.4809	-
1.0215	8820	1.5129	-
1.0216	8830	1.508	-
1.0217	8840	1.4991	-
1.0218	8850	1.5162	-
1.0219	8860	1.4969	-
1.0220	8870	1.5061	-
1.0221	8880	1.5091	-
1.0222	8890	1.4782	-
1.0223	8900	1.519	0.5867
1.0224	8910	1.5132	-
1.0225	8920	1.4933	-
1.0226	8930	1.5157	-
1.0227	8940	1.4935	-
1.0228	8950	1.4979	-
1.0229	8960	1.5165	-
1.0230	8970	1.497	-
1.0231	8980	1.5104	-
1.0232	8990	1.5089	-
1.0233	9000	1.5171	0.5962
1.0234	9010	1.5238	-
1.0235	9020	1.4794	-
1.0236	9030	1.5047	-
1.0237	9040	1.4857	-
1.0238	9050	1.524	-
1.0239	9060	1.5167	-
1.0240	9070	1.5014	-
1.0241	9080	1.5148	-
1.0242	9090	1.5196	-
1.0243	9100	1.5058	0.5872
1.0244	9110	1.4701	-
1.0245	9120	1.5206	-
1.0246	9130	1.5175	-
1.0247	9140	1.4931	-
1.0248	9150	1.507	-
1.0249	9160	1.5117	-
1.0250	9170	1.5087	-
1.0251	9180	1.5223	-
1.0252	9190	1.5108	-
1.0253	9200	1.4735	0.5923
1.0254	9210	1.5003	-
1.0255	9220	1.4958	-
1.0256	9230	1.5119	-
1.0257	9240	1.5167	-
1.0258	9250	1.5157	-
1.0259	9260	1.4788	-
1.0260	9270	1.4943	-
1.0261	9280	1.5019	-
1.0262	9290	1.5167	-
1.0263	9300	1.5088	0.5915
1.0264	9310	1.5033	-
1.0265	9320	1.5063	-
1.0266	9330	1.4833	-
1.0267	9340	1.4704	-
1.0268	9350	1.4997	-
1.0269	9360	1.4968	-
1.0270	9370	1.4838	-
1.0271	9380	1.4917	-
1.0272	9390	1.5128	-
1.0273	9400	1.4995	0.5876
1.0274	9410	1.4883	-
1.0275	9420	1.5139	-
1.0276	9430	1.5102	-
1.0277	9440	1.5045	-
1.0278	9450	1.5055	-
1.0279	9460	1.5116	-
1.0280	9470	1.509	-
1.0281	9480	1.5028	-
1.0282	9490	1.495	-
1.0283	9500	1.5127	0.5912
1.0284	9510	1.5037	-
1.0285	9520	1.4967	-
1.0286	9530	1.5034	-
1.0287	9540	1.498	-
1.0288	9550	1.4735	-
1.0289	9560	1.4879	-
1.0290	9570	1.501	-
1.0291	9580	1.4822	-
1.0292	9590	1.5011	-
1.0293	9600	1.484	0.5868
1.0294	9610	1.4973	-
1.0295	9620	1.5193	-
1.0296	9630	1.4968	-
1.0297	9640	1.5089	-
1.0298	9650	1.485	-
1.0299	9660	1.5018	-
1.0300	9670	1.5024	-
1.0301	9680	1.4949	-
1.0302	9690	1.5043	-
1.0303	9700	1.5094	0.5894
1.0304	9710	1.4973	-
1.0305	9720	1.4851	-
1.0306	9730	1.4908	-
1.0307	9740	1.5076	-
1.0308	9750	1.4904	-
1.0309	9760	1.4939	-
1.0310	9770	1.4837	-
1.0311	9780	1.5059	-
1.0312	9790	1.5027	-
1.0313	9800	1.4951	0.5883
1.0314	9810	1.511	-
1.0315	9820	1.5063	-
1.0316	9830	1.5104	-
1.0317	9840	1.4894	-
1.0318	9850	1.5114	-
1.0319	9860	1.5072	-
1.0320	9870	1.4926	-
1.0321	9880	1.4837	-
1.0322	9890	1.4932	-
1.0323	9900	1.5093	0.5908
1.0324	9910	1.4866	-
1.0325	9920	1.4943	-
1.0326	9930	1.4893	-
1.0327	9940	1.5086	-
1.0328	9950	1.4815	-
1.0329	9960	1.5054	-
1.0330	9970	1.4754	-
1.0331	9980	1.4984	-
1.0332	9990	1.4948	-
1.0333	10000	1.5052	0.5932

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}
}