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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("reasonwang/embedding-qwen3-1.7b-embedding_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, 0.8993, 0.7656],
#         [0.8993, 1.0000, 0.6804],
#         [0.7656, 0.6804, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: validation_retrieval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.62
cosine_accuracy@3	0.79
cosine_accuracy@5	0.88
cosine_accuracy@10	0.93
cosine_precision@1	0.62
cosine_precision@3	0.4733
cosine_precision@5	0.406
cosine_precision@10	0.292
cosine_recall@1	0.1206
cosine_recall@3	0.2301
cosine_recall@5	0.3083
cosine_recall@10	0.3879
cosine_ndcg@10	0.4788
cosine_ndcg@100	0.551
cosine_mrr@10	0.7219
cosine_mrr@100	0.725
cosine_map@100	0.3511

Training Details

Training Dataset

generator

Dataset: generator
Columns: sentence1 and sentence2
Loss: CachedMultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "mini_batch_size": 4, 5 "gather_across_devices": false 6}

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	5.2053	-
0.0001	10	3.8468	-
0.0002	20	2.8407	-
0.0003	30	2.5497	-
0.0004	40	2.4161	-
0.0005	50	2.3331	-
0.0006	60	2.3184	-
0.0007	70	2.2386	-
0.0008	80	2.2135	-
0.0009	90	2.1836	-
0.001	100	2.1797	0.4941
0.0011	110	2.1426	-
0.0012	120	2.1384	-
0.0013	130	2.1306	-
0.0014	140	2.1141	-
0.0015	150	2.1241	-
0.0016	160	2.0885	-
0.0017	170	2.0815	-
0.0018	180	2.099	-
0.0019	190	2.0519	-
0.002	200	2.0482	0.5110
0.0021	210	2.0405	-
0.0022	220	2.0122	-
0.0023	230	2.025	-
0.0024	240	2.011	-
0.0025	250	2.0368	-
0.0026	260	1.9879	-
0.0027	270	2.0191	-
0.0028	280	1.9754	-
0.0029	290	2.0114	-
0.003	300	1.9769	0.5158
0.0031	310	1.9594	-
0.0032	320	1.9792	-
0.0033	330	1.9615	-
0.0034	340	1.955	-
0.0035	350	1.9725	-
0.0036	360	1.9524	-
0.0037	370	1.9435	-
0.0038	380	1.9398	-
0.0039	390	1.9251	-
0.004	400	1.8997	0.5228
0.0041	410	1.9296	-
0.0042	420	1.9267	-
0.0043	430	1.9301	-
0.0044	440	1.9417	-
0.0045	450	1.9179	-
0.0046	460	1.9091	-
0.0047	470	1.9147	-
0.0048	480	1.9187	-
0.0049	490	1.9345	-
0.005	500	1.883	0.5258
0.0051	510	1.9023	-
0.0052	520	1.8664	-
0.0053	530	1.9142	-
0.0054	540	1.8902	-
0.0055	550	1.877	-
0.0056	560	1.8753	-
0.0057	570	1.8804	-
0.0058	580	1.8923	-
0.0059	590	1.8501	-
0.006	600	1.8499	0.5265
0.0061	610	1.8719	-
0.0062	620	1.8692	-
0.0063	630	1.8813	-
0.0064	640	1.8758	-
0.0065	650	1.8721	-
0.0066	660	1.8466	-
0.0067	670	1.8873	-
0.0068	680	1.8429	-
0.0069	690	1.8719	-
0.007	700	1.8478	0.5271
0.0071	710	1.8616	-
0.0072	720	1.8486	-
0.0073	730	1.8766	-
0.0074	740	1.8614	-
0.0075	750	1.8722	-
0.0076	760	1.829	-
0.0077	770	1.8341	-
0.0078	780	1.8304	-
0.0079	790	1.8494	-
0.008	800	1.838	0.5328
0.0081	810	1.851	-
0.0082	820	1.8359	-
0.0083	830	1.8528	-
0.0084	840	1.8295	-
0.0085	850	1.8337	-
0.0086	860	1.8072	-
0.0087	870	1.8068	-
0.0088	880	1.8102	-
0.0089	890	1.8199	-
0.009	900	1.8308	0.5324
0.0091	910	1.8155	-
0.0092	920	1.7918	-
0.0093	930	1.7969	-
0.0094	940	1.8054	-
0.0095	950	1.8032	-
0.0096	960	1.7972	-
0.0097	970	1.8063	-
0.0098	980	1.8227	-
0.0099	990	1.8111	-
0.01	1000	1.7909	0.5312
0.0101	1010	1.8105	-
0.0102	1020	1.7976	-
0.0103	1030	1.8117	-
0.0104	1040	1.7823	-
0.0105	1050	1.7967	-
0.0106	1060	1.7941	-
0.0107	1070	1.8054	-
0.0108	1080	1.7938	-
0.0109	1090	1.784	-
0.011	1100	1.8	0.5342
0.0111	1110	1.7895	-
0.0112	1120	1.805	-
0.0113	1130	1.8063	-
0.0114	1140	1.8011	-
0.0115	1150	1.7609	-
0.0116	1160	1.7658	-
0.0117	1170	1.7393	-
0.0118	1180	1.7716	-
0.0119	1190	1.7546	-
0.012	1200	1.776	0.5351
0.0121	1210	1.7735	-
0.0122	1220	1.7815	-
0.0123	1230	1.7805	-
0.0124	1240	1.7677	-
0.0125	1250	1.7697	-
0.0126	1260	1.7599	-
0.0127	1270	1.752	-
0.0128	1280	1.7591	-
0.0129	1290	1.7519	-
0.013	1300	1.7782	0.5344
0.0131	1310	1.741	-
0.0132	1320	1.7475	-
0.0133	1330	1.7903	-
0.0134	1340	1.7488	-
0.0135	1350	1.7426	-
0.0136	1360	1.7589	-
0.0137	1370	1.7327	-
0.0138	1380	1.7453	-
0.0139	1390	1.7453	-
0.014	1400	1.7347	0.5345
0.0141	1410	1.7467	-
0.0142	1420	1.743	-
0.0143	1430	1.7587	-
0.0144	1440	1.73	-
0.0145	1450	1.7453	-
0.0146	1460	1.7387	-
0.0147	1470	1.7431	-
0.0148	1480	1.7542	-
0.0149	1490	1.764	-
0.015	1500	1.7297	0.5376
0.0151	1510	1.7404	-
0.0152	1520	1.7497	-
0.0153	1530	1.756	-
0.0154	1540	1.7426	-
0.0155	1550	1.7246	-
0.0156	1560	1.7217	-
0.0157	1570	1.7319	-
0.0158	1580	1.7642	-
0.0159	1590	1.7345	-
0.016	1600	1.7312	0.5360
0.0161	1610	1.7394	-
0.0162	1620	1.7025	-
0.0163	1630	1.7431	-
0.0164	1640	1.7216	-
0.0165	1650	1.7378	-
0.0166	1660	1.7217	-
0.0167	1670	1.7424	-
0.0168	1680	1.7349	-
0.0169	1690	1.7191	-
0.017	1700	1.7288	0.5380
0.0171	1710	1.7173	-
0.0172	1720	1.7215	-
0.0173	1730	1.7108	-
0.0174	1740	1.6972	-
0.0175	1750	1.7177	-
0.0176	1760	1.7259	-
0.0177	1770	1.7103	-
0.0178	1780	1.7286	-
0.0179	1790	1.7127	-
0.018	1800	1.7102	0.5373
0.0181	1810	1.6951	-
0.0182	1820	1.7221	-
0.0183	1830	1.7135	-
0.0184	1840	1.6976	-
0.0185	1850	1.719	-
0.0186	1860	1.7182	-
0.0187	1870	1.7	-
0.0188	1880	1.7307	-
0.0189	1890	1.7024	-
0.019	1900	1.7092	0.5395
0.0191	1910	1.6905	-
0.0192	1920	1.7068	-
0.0193	1930	1.7053	-
0.0194	1940	1.7104	-
0.0195	1950	1.7068	-
0.0196	1960	1.7136	-
0.0197	1970	1.672	-
0.0198	1980	1.7229	-
0.0199	1990	1.6988	-
0.02	2000	1.7009	0.5376
0.0201	2010	1.6763	-
0.0202	2020	1.7004	-
0.0203	2030	1.6916	-
0.0204	2040	1.7155	-
0.0205	2050	1.7055	-
0.0206	2060	1.6892	-
0.0207	2070	1.7236	-
0.0208	2080	1.6971	-
0.0209	2090	1.7105	-
0.021	2100	1.6989	0.5398
0.0211	2110	1.7052	-
0.0212	2120	1.6936	-
0.0213	2130	1.7055	-
0.0214	2140	1.7038	-
0.0215	2150	1.6902	-
0.0216	2160	1.6808	-
0.0217	2170	1.6859	-
0.0218	2180	1.7133	-
0.0219	2190	1.6992	-
0.022	2200	1.6929	0.5389
0.0221	2210	1.6918	-
0.0222	2220	1.6732	-
0.0223	2230	1.7048	-
0.0224	2240	1.6682	-
0.0225	2250	1.6733	-
0.0226	2260	1.6655	-
0.0227	2270	1.6712	-
0.0228	2280	1.6707	-
0.0229	2290	1.668	-
0.023	2300	1.7152	0.5413
0.0231	2310	1.6816	-
0.0232	2320	1.6642	-
0.0233	2330	1.6809	-
0.0234	2340	1.6925	-
0.0235	2350	1.679	-
0.0236	2360	1.6857	-
0.0237	2370	1.6958	-
0.0238	2380	1.6943	-
0.0239	2390	1.6487	-
0.024	2400	1.6818	0.5352
0.0241	2410	1.6752	-
0.0242	2420	1.6707	-
0.0243	2430	1.6891	-
0.0244	2440	1.6691	-
0.0245	2450	1.6786	-
0.0246	2460	1.6831	-
0.0247	2470	1.6788	-
0.0248	2480	1.69	-
0.0249	2490	1.6811	-
0.025	2500	1.6584	0.5425
0.0251	2510	1.6523	-
0.0252	2520	1.6947	-
0.0253	2530	1.6542	-
0.0254	2540	1.6411	-
0.0255	2550	1.6729	-
0.0256	2560	1.6851	-
0.0257	2570	1.6642	-
0.0258	2580	1.6614	-
0.0259	2590	1.6684	-
0.026	2600	1.6676	0.5402
0.0261	2610	1.6678	-
0.0262	2620	1.6741	-
0.0263	2630	1.6769	-
0.0264	2640	1.6468	-
0.0265	2650	1.6688	-
0.0266	2660	1.6543	-
0.0267	2670	1.7078	-
0.0268	2680	1.6353	-
0.0269	2690	1.665	-
0.027	2700	1.6629	0.5441
0.0271	2710	1.6567	-
0.0272	2720	1.6395	-
0.0273	2730	1.6629	-
0.0274	2740	1.6852	-
0.0275	2750	1.6635	-
0.0276	2760	1.6765	-
0.0277	2770	1.6598	-
0.0278	2780	1.6547	-
0.0279	2790	1.6741	-
0.028	2800	1.6664	0.5423
0.0281	2810	1.6446	-
0.0282	2820	1.6662	-
0.0283	2830	1.6566	-
0.0284	2840	1.6397	-
0.0285	2850	1.6728	-
0.0286	2860	1.6592	-
0.0287	2870	1.657	-
0.0288	2880	1.6498	-
0.0289	2890	1.6375	-
0.029	2900	1.6437	0.5376
0.0291	2910	1.6309	-
0.0292	2920	1.6665	-
0.0293	2930	1.6666	-
0.0294	2940	1.6476	-
0.0295	2950	1.6604	-
0.0296	2960	1.6595	-
0.0297	2970	1.6476	-
0.0298	2980	1.6716	-
0.0299	2990	1.629	-
0.03	3000	1.6383	0.5425
0.0301	3010	1.6545	-
0.0302	3020	1.6477	-
0.0303	3030	1.6137	-
0.0304	3040	1.6581	-
0.0305	3050	1.6696	-
0.0306	3060	1.6316	-
0.0307	3070	1.6439	-
0.0308	3080	1.6537	-
0.0309	3090	1.6466	-
0.031	3100	1.6388	0.5416
0.0311	3110	1.6467	-
0.0312	3120	1.6235	-
0.0313	3130	1.6201	-
0.0314	3140	1.6342	-
0.0315	3150	1.6487	-
0.0316	3160	1.6531	-
0.0317	3170	1.6552	-
0.0318	3180	1.65	-
0.0319	3190	1.6457	-
0.032	3200	1.6448	0.5427
0.0321	3210	1.641	-
0.0322	3220	1.6612	-
0.0323	3230	1.6482	-
0.0324	3240	1.6362	-
0.0325	3250	1.6299	-
0.0326	3260	1.642	-
0.0327	3270	1.6443	-
0.0328	3280	1.6362	-
0.0329	3290	1.6234	-
0.033	3300	1.6294	0.5409
0.0331	3310	1.6278	-
0.0332	3320	1.6201	-
0.0333	3330	1.6059	-
0.0334	3340	1.6609	-
0.0335	3350	1.6263	-
0.0336	3360	1.6113	-
0.0337	3370	1.6403	-
0.0338	3380	1.6425	-
0.0339	3390	1.6274	-
0.034	3400	1.6364	0.5377
0.0341	3410	1.6252	-
0.0342	3420	1.6144	-
0.0343	3430	1.6334	-
0.0344	3440	1.6172	-
0.0345	3450	1.6293	-
0.0346	3460	1.6308	-
0.0347	3470	1.6309	-
0.0348	3480	1.6178	-
0.0349	3490	1.6426	-
0.035	3500	1.6178	0.5466
0.0351	3510	1.6516	-
0.0352	3520	1.6249	-
0.0353	3530	1.6355	-
0.0354	3540	1.6451	-
0.0355	3550	1.611	-
0.0356	3560	1.6265	-
0.0357	3570	1.633	-
0.0358	3580	1.6222	-
0.0359	3590	1.616	-
0.036	3600	1.6226	0.5412
0.0361	3610	1.6251	-
0.0362	3620	1.6148	-
0.0363	3630	1.6348	-
0.0364	3640	1.6421	-
0.0365	3650	1.5937	-
0.0366	3660	1.6469	-
0.0367	3670	1.6259	-
0.0368	3680	1.6216	-
0.0369	3690	1.6206	-
0.037	3700	1.6159	0.5422
0.0371	3710	1.6185	-
0.0372	3720	1.6414	-
0.0373	3730	1.6179	-
0.0374	3740	1.6002	-
0.0375	3750	1.6226	-
0.0376	3760	1.6305	-
0.0377	3770	1.6198	-
0.0378	3780	1.6184	-
0.0379	3790	1.6445	-
0.038	3800	1.6291	0.5396
0.0381	3810	1.6029	-
0.0382	3820	1.6039	-
0.0383	3830	1.6351	-
0.0384	3840	1.6238	-
0.0385	3850	1.6086	-
0.0386	3860	1.6435	-
0.0387	3870	1.5971	-
0.0388	3880	1.6114	-
0.0389	3890	1.6077	-
0.039	3900	1.584	0.5387
0.0391	3910	1.6086	-
0.0392	3920	1.6165	-
0.0393	3930	1.611	-
0.0394	3940	1.6102	-
0.0395	3950	1.5823	-
0.0396	3960	1.6146	-
0.0397	3970	1.5876	-
0.0398	3980	1.605	-
0.0399	3990	1.629	-
0.04	4000	1.6349	0.5434
0.0401	4010	1.6273	-
0.0402	4020	1.6252	-
0.0403	4030	1.6426	-
0.0404	4040	1.6003	-
0.0405	4050	1.6135	-
0.0406	4060	1.5962	-
0.0407	4070	1.6133	-
0.0408	4080	1.5951	-
0.0409	4090	1.6001	-
0.041	4100	1.5997	0.5432
0.0411	4110	1.6117	-
0.0412	4120	1.6178	-
0.0413	4130	1.6151	-
0.0414	4140	1.6253	-
0.0415	4150	1.613	-
0.0416	4160	1.6197	-
0.0417	4170	1.5976	-
0.0418	4180	1.6213	-
0.0419	4190	1.5855	-
0.042	4200	1.6117	0.5443
0.0421	4210	1.6309	-
0.0422	4220	1.6047	-
0.0423	4230	1.6154	-
0.0424	4240	1.6001	-
0.0425	4250	1.619	-
0.0426	4260	1.6137	-
0.0427	4270	1.5896	-
0.0428	4280	1.622	-
0.0429	4290	1.6047	-
0.043	4300	1.6177	0.5408
0.0431	4310	1.6021	-
0.0432	4320	1.6219	-
0.0433	4330	1.5891	-
0.0434	4340	1.6268	-
0.0435	4350	1.6135	-
0.0436	4360	1.5985	-
0.0437	4370	1.6017	-
0.0438	4380	1.6061	-
0.0439	4390	1.619	-
0.044	4400	1.5897	0.5449
0.0441	4410	1.6126	-
0.0442	4420	1.6143	-
0.0443	4430	1.6213	-
0.0444	4440	1.6026	-
0.0445	4450	1.5946	-
0.0446	4460	1.6191	-
0.0447	4470	1.5861	-
0.0448	4480	1.6075	-
0.0449	4490	1.5854	-
0.045	4500	1.5773	0.5399
0.0451	4510	1.6214	-
0.0452	4520	1.6005	-
0.0453	4530	1.579	-
0.0454	4540	1.6039	-
0.0455	4550	1.5778	-
0.0456	4560	1.6141	-
0.0457	4570	1.6088	-
0.0458	4580	1.5981	-
0.0459	4590	1.5863	-
0.046	4600	1.5873	0.5438
0.0461	4610	1.595	-
0.0462	4620	1.5976	-
0.0463	4630	1.5772	-
0.0464	4640	1.5724	-
0.0465	4650	1.5905	-
0.0466	4660	1.6146	-
0.0467	4670	1.6097	-
0.0468	4680	1.6069	-
0.0469	4690	1.5915	-
0.047	4700	1.5934	0.5478
0.0471	4710	1.6011	-
0.0472	4720	1.6017	-
0.0473	4730	1.6104	-
0.0474	4740	1.5903	-
0.0475	4750	1.6011	-
0.0476	4760	1.5801	-
0.0477	4770	1.5784	-
0.0478	4780	1.6015	-
0.0479	4790	1.591	-
0.048	4800	1.6159	0.5480
0.0481	4810	1.5994	-
0.0482	4820	1.5687	-
0.0483	4830	1.5986	-
0.0484	4840	1.6035	-
0.0485	4850	1.5855	-
0.0486	4860	1.5647	-
0.0487	4870	1.602	-
0.0488	4880	1.5805	-
0.0489	4890	1.5876	-
0.049	4900	1.6101	0.5447
0.0491	4910	1.5817	-
0.0492	4920	1.5869	-
0.0493	4930	1.5883	-
0.0494	4940	1.5646	-
0.0495	4950	1.571	-
0.0496	4960	1.5814	-
0.0497	4970	1.5863	-
0.0498	4980	1.5599	-
0.0499	4990	1.6061	-
0.05	5000	1.5672	0.5471
0.0501	5010	1.5785	-
0.0502	5020	1.5813	-
0.0503	5030	1.5932	-
0.0504	5040	1.5807	-
0.0505	5050	1.5704	-
0.0506	5060	1.5986	-
0.0507	5070	1.5812	-
0.0508	5080	1.5922	-
0.0509	5090	1.5789	-
0.051	5100	1.5829	0.5490
0.0511	5110	1.5634	-
0.0512	5120	1.5989	-
0.0513	5130	1.5928	-
0.0514	5140	1.5686	-
0.0515	5150	1.5548	-
0.0516	5160	1.599	-
0.0517	5170	1.5904	-
0.0518	5180	1.5898	-
0.0519	5190	1.578	-
0.052	5200	1.5721	0.5484
0.0521	5210	1.5729	-
0.0522	5220	1.5905	-
0.0523	5230	1.5629	-
0.0524	5240	1.5498	-
0.0525	5250	1.577	-
0.0526	5260	1.5995	-
0.0527	5270	1.5591	-
0.0528	5280	1.5772	-
0.0529	5290	1.5927	-
0.053	5300	1.5812	0.5467
0.0531	5310	1.5703	-
0.0532	5320	1.5885	-
0.0533	5330	1.5879	-
0.0534	5340	1.5884	-
0.0535	5350	1.5692	-
0.0536	5360	1.6	-
0.0537	5370	1.5907	-
0.0538	5380	1.5728	-
0.0539	5390	1.5876	-
0.054	5400	1.5661	0.5485
0.0541	5410	1.5525	-
0.0542	5420	1.5855	-
0.0543	5430	1.5752	-
0.0544	5440	1.5785	-
0.0545	5450	1.5825	-
0.0546	5460	1.5828	-
0.0547	5470	1.5743	-
0.0548	5480	1.5873	-
0.0549	5490	1.5909	-
0.055	5500	1.5811	0.5538
0.0551	5510	1.5871	-
0.0552	5520	1.5569	-
0.0553	5530	1.5748	-
0.0554	5540	1.6094	-
0.0555	5550	1.5623	-
0.0556	5560	1.5732	-
0.0557	5570	1.5724	-
0.0558	5580	1.5731	-
0.0559	5590	1.5873	-
0.056	5600	1.5913	0.5512
0.0561	5610	1.5916	-
0.0562	5620	1.5684	-
0.0563	5630	1.5526	-
0.0564	5640	1.5699	-
0.0565	5650	1.5732	-
0.0566	5660	1.5797	-
0.0567	5670	1.5723	-
0.0568	5680	1.5719	-
0.0569	5690	1.572	-
0.057	5700	1.5484	0.5491
0.0571	5710	1.5927	-
0.0572	5720	1.5614	-
0.0573	5730	1.5723	-
0.0574	5740	1.5675	-
0.0575	5750	1.5577	-
0.0576	5760	1.5801	-
0.0577	5770	1.5685	-
0.0578	5780	1.5802	-
0.0579	5790	1.5707	-
0.058	5800	1.588	0.5508
0.0581	5810	1.5642	-
0.0582	5820	1.5649	-
0.0583	5830	1.5724	-
0.0584	5840	1.5701	-
0.0585	5850	1.5731	-
0.0586	5860	1.5625	-
0.0587	5870	1.5822	-
0.0588	5880	1.5668	-
0.0589	5890	1.5742	-
0.059	5900	1.5413	0.5498
0.0591	5910	1.5535	-
0.0592	5920	1.5653	-
0.0593	5930	1.5604	-
0.0594	5940	1.5905	-
0.0595	5950	1.5689	-
0.0596	5960	1.5578	-
0.0597	5970	1.5616	-
0.0598	5980	1.5945	-
0.0599	5990	1.5721	-
0.06	6000	1.5701	0.5506
0.0601	6010	1.5486	-
0.0602	6020	1.5581	-
0.0603	6030	1.5801	-
0.0604	6040	1.5684	-
0.0605	6050	1.583	-
0.0606	6060	1.5885	-
0.0607	6070	1.5647	-
0.0608	6080	1.5826	-
0.0609	6090	1.5858	-
0.061	6100	1.5618	0.5501
0.0611	6110	1.5538	-
0.0612	6120	1.5721	-
0.0613	6130	1.5919	-
0.0614	6140	1.5693	-
0.0615	6150	1.5578	-
0.0616	6160	1.552	-
0.0617	6170	1.5584	-
0.0618	6180	1.5455	-
0.0619	6190	1.5421	-
0.062	6200	1.5551	0.5552
0.0621	6210	1.5706	-
0.0622	6220	1.5732	-
0.0623	6230	1.5619	-
0.0624	6240	1.5732	-
0.0625	6250	1.5554	-
0.0626	6260	1.5734	-
0.0627	6270	1.5747	-
0.0628	6280	1.571	-
0.0629	6290	1.5616	-
0.063	6300	1.5575	0.5546
0.0631	6310	1.5586	-
0.0632	6320	1.5857	-
0.0633	6330	1.5593	-
0.0634	6340	1.556	-
0.0635	6350	1.5558	-
0.0636	6360	1.5799	-
0.0637	6370	1.5772	-
0.0638	6380	1.5477	-
0.0639	6390	1.5823	-
0.064	6400	1.5462	0.5497
0.0641	6410	1.57	-
0.0642	6420	1.5551	-
0.0643	6430	1.5687	-
0.0644	6440	1.5703	-
0.0645	6450	1.579	-
0.0646	6460	1.5426	-
0.0647	6470	1.5816	-
0.0648	6480	1.5712	-
0.0649	6490	1.5746	-
0.065	6500	1.5648	0.5503
0.0651	6510	1.5443	-
0.0652	6520	1.5499	-
0.0653	6530	1.5648	-
0.0654	6540	1.5585	-
0.0655	6550	1.5694	-
0.0656	6560	1.5579	-
0.0657	6570	1.5762	-
0.0658	6580	1.5494	-
0.0659	6590	1.5752	-
0.066	6600	1.554	0.5493
0.0661	6610	1.5469	-
0.0662	6620	1.5532	-
0.0663	6630	1.535	-
0.0664	6640	1.5146	-
0.0665	6650	1.5528	-
0.0666	6660	1.5626	-
0.0667	6670	1.5476	-
0.0668	6680	1.5594	-
0.0669	6690	1.5441	-
0.067	6700	1.562	0.5504
0.0671	6710	1.5468	-
0.0672	6720	1.5462	-
0.0673	6730	1.5711	-
0.0674	6740	1.5659	-
0.0675	6750	1.5413	-
0.0676	6760	1.5497	-
0.0677	6770	1.5477	-
0.0678	6780	1.548	-
0.0679	6790	1.5639	-
0.068	6800	1.5676	0.5500
0.0681	6810	1.5647	-
0.0682	6820	1.5663	-
0.0683	6830	1.5462	-
0.0684	6840	1.5675	-
0.0685	6850	1.5563	-
0.0686	6860	1.5754	-
0.0687	6870	1.5513	-
0.0688	6880	1.5629	-
0.0689	6890	1.5558	-
0.069	6900	1.5536	0.5476
0.0691	6910	1.5618	-
0.0692	6920	1.5559	-
0.0693	6930	1.5661	-
0.0694	6940	1.5554	-
0.0695	6950	1.5501	-
0.0696	6960	1.581	-
0.0697	6970	1.5303	-
0.0698	6980	1.5432	-
0.0699	6990	1.5486	-
0.07	7000	1.5517	0.5495
0.0701	7010	1.5765	-
0.0702	7020	1.5536	-
0.0703	7030	1.5425	-
0.0704	7040	1.5593	-
0.0705	7050	1.548	-
0.0706	7060	1.5501	-
0.0707	7070	1.5331	-
0.0708	7080	1.567	-
0.0709	7090	1.5906	-
0.071	7100	1.5586	0.5474
0.0711	7110	1.5468	-
0.0712	7120	1.5503	-
0.0713	7130	1.5514	-
0.0714	7140	1.5584	-
0.0715	7150	1.5471	-
0.0716	7160	1.5462	-
0.0717	7170	1.5498	-
0.0718	7180	1.558	-
0.0719	7190	1.5495	-
0.072	7200	1.5585	0.5486
0.0721	7210	1.536	-
0.0722	7220	1.5493	-
0.0723	7230	1.5442	-
0.0724	7240	1.5409	-
0.0725	7250	1.536	-
0.0726	7260	1.5674	-
0.0727	7270	1.5595	-
0.0728	7280	1.5184	-
0.0729	7290	1.5694	-
0.073	7300	1.5593	0.5513
0.0731	7310	1.5766	-
0.0732	7320	1.5562	-
0.0733	7330	1.5872	-
0.0734	7340	1.5705	-
0.0735	7350	1.5622	-
0.0736	7360	1.5577	-
0.0737	7370	1.5422	-
0.0738	7380	1.5643	-
0.0739	7390	1.5713	-
0.074	7400	1.533	0.5500
0.0741	7410	1.5204	-
0.0742	7420	1.5445	-
0.0743	7430	1.5503	-
0.0744	7440	1.5243	-
0.0745	7450	1.5465	-
0.0746	7460	1.559	-
0.0747	7470	1.5493	-
0.0748	7480	1.5589	-
0.0749	7490	1.5637	-
0.075	7500	1.5547	0.5484
0.0751	7510	1.5296	-
0.0752	7520	1.5466	-
0.0753	7530	1.5259	-
0.0754	7540	1.5496	-
0.0755	7550	1.5273	-
0.0756	7560	1.5606	-
0.0757	7570	1.5408	-
0.0758	7580	1.5529	-
0.0759	7590	1.5481	-
0.076	7600	1.5571	0.5541
0.0761	7610	1.5322	-
0.0762	7620	1.567	-
0.0763	7630	1.5387	-
0.0764	7640	1.554	-
0.0765	7650	1.5284	-
0.0766	7660	1.5255	-
0.0767	7670	1.5421	-
0.0768	7680	1.5538	-
0.0769	7690	1.5309	-
0.077	7700	1.549	0.5519
0.0771	7710	1.5373	-
0.0772	7720	1.5315	-
0.0773	7730	1.5345	-
0.0774	7740	1.56	-
0.0775	7750	1.5678	-
0.0776	7760	1.5653	-
0.0777	7770	1.521	-
0.0778	7780	1.5377	-
0.0779	7790	1.5518	-
0.078	7800	1.5454	0.5490
0.0781	7810	1.5227	-
0.0782	7820	1.5604	-
0.0783	7830	1.5283	-
0.0784	7840	1.5448	-
0.0785	7850	1.5116	-
0.0786	7860	1.5223	-
0.0787	7870	1.5497	-
0.0788	7880	1.5417	-
0.0789	7890	1.5358	-
0.079	7900	1.5504	0.5544
0.0791	7910	1.516	-
0.0792	7920	1.5422	-
0.0793	7930	1.537	-
0.0794	7940	1.5479	-
0.0795	7950	1.5579	-
0.0796	7960	1.5369	-
0.0797	7970	1.5321	-
0.0798	7980	1.522	-
0.0799	7990	1.5389	-
0.08	8000	1.5274	0.5526
0.0801	8010	1.5537	-
0.0802	8020	1.5397	-
0.0803	8030	1.5671	-
0.0804	8040	1.5349	-
0.0805	8050	1.5407	-
0.0806	8060	1.5563	-
0.0807	8070	1.5581	-
0.0808	8080	1.5423	-
0.0809	8090	1.5148	-
0.081	8100	1.5557	0.5544
0.0811	8110	1.5404	-
0.0812	8120	1.5368	-
0.0813	8130	1.5161	-
0.0814	8140	1.5595	-
0.0815	8150	1.5493	-
0.0816	8160	1.5312	-
0.0817	8170	1.5326	-
0.0818	8180	1.5424	-
0.0819	8190	1.5325	-
0.082	8200	1.5458	0.5561
0.0821	8210	1.5397	-
0.0822	8220	1.5438	-
0.0823	8230	1.5237	-
0.0824	8240	1.5396	-
0.0825	8250	1.5365	-
0.0826	8260	1.5609	-
0.0827	8270	1.533	-
0.0828	8280	1.5367	-
0.0829	8290	1.5316	-
0.083	8300	1.5386	0.5528
0.0831	8310	1.5259	-
0.0832	8320	1.5205	-
0.0833	8330	1.5561	-
0.0834	8340	1.533	-
0.0835	8350	1.5684	-
0.0836	8360	1.5475	-
0.0837	8370	1.5195	-
0.0838	8380	1.5388	-
0.0839	8390	1.564	-
0.084	8400	1.5572	0.5490
0.0841	8410	1.5567	-
0.0842	8420	1.5383	-
0.0843	8430	1.5645	-
0.0844	8440	1.5499	-
0.0845	8450	1.5267	-
0.0846	8460	1.5538	-
0.0847	8470	1.5635	-
0.0848	8480	1.5365	-
0.0849	8490	1.5374	-
0.085	8500	1.5453	0.5507
0.0851	8510	1.5155	-
0.0852	8520	1.5505	-
0.0853	8530	1.5381	-
0.0854	8540	1.5337	-
0.0855	8550	1.5475	-
0.0856	8560	1.5421	-
0.0857	8570	1.5318	-
0.0858	8580	1.5404	-
0.0859	8590	1.5227	-
0.086	8600	1.5323	0.5498
0.0861	8610	1.5245	-
0.0862	8620	1.5435	-
0.0863	8630	1.5516	-
0.0864	8640	1.5394	-
0.0865	8650	1.5141	-
0.0866	8660	1.5289	-
0.0867	8670	1.5191	-
0.0868	8680	1.5349	-
0.0869	8690	1.5507	-
0.087	8700	1.5337	0.5532
0.0871	8710	1.5471	-
0.0872	8720	1.5267	-
0.0873	8730	1.5308	-
0.0874	8740	1.5576	-
0.0875	8750	1.5424	-
0.0876	8760	1.5518	-
0.0877	8770	1.5316	-
0.0878	8780	1.5369	-
0.0879	8790	1.5412	-
0.088	8800	1.5407	0.5487
0.0881	8810	1.5257	-
0.0882	8820	1.5318	-
0.0883	8830	1.5214	-
0.0884	8840	1.5321	-
0.0885	8850	1.5282	-
0.0886	8860	1.5262	-
0.0887	8870	1.5545	-
0.0888	8880	1.5407	-
0.0889	8890	1.564	-
0.089	8900	1.5287	0.5518
0.0891	8910	1.5353	-
0.0892	8920	1.5155	-
0.0893	8930	1.5416	-
0.0894	8940	1.546	-
0.0895	8950	1.5349	-
0.0896	8960	1.5203	-
0.0897	8970	1.5282	-
0.0898	8980	1.5111	-
0.0899	8990	1.5121	-
0.09	9000	1.5209	0.5519
0.0901	9010	1.5333	-
0.0902	9020	1.5305	-
0.0903	9030	1.5397	-
0.0904	9040	1.523	-
0.0905	9050	1.5446	-
0.0906	9060	1.5378	-
0.0907	9070	1.533	-
0.0908	9080	1.5271	-
0.0909	9090	1.5201	-
0.091	9100	1.526	0.5524
0.0911	9110	1.5307	-
0.0912	9120	1.572	-
0.0913	9130	1.5016	-
0.0914	9140	1.526	-
0.0915	9150	1.5326	-
0.0916	9160	1.5189	-
0.0917	9170	1.5298	-
0.0918	9180	1.5211	-
0.0919	9190	1.5237	-
0.092	9200	1.5121	0.5497
0.0921	9210	1.4938	-
0.0922	9220	1.5094	-
0.0923	9230	1.5265	-
0.0924	9240	1.5278	-
0.0925	9250	1.5255	-
0.0926	9260	1.4975	-
0.0927	9270	1.5117	-
0.0928	9280	1.5378	-
0.0929	9290	1.5248	-
0.093	9300	1.5222	0.5531
0.0931	9310	1.5056	-
0.0932	9320	1.5361	-
0.0933	9330	1.5426	-
0.0934	9340	1.5023	-
0.0935	9350	1.5056	-
0.0936	9360	1.5058	-
0.0937	9370	1.5299	-
0.0938	9380	1.5178	-
0.0939	9390	1.532	-
0.094	9400	1.5248	0.5577
0.0941	9410	1.5374	-
0.0942	9420	1.518	-
0.0943	9430	1.5299	-
0.0944	9440	1.5432	-
0.0945	9450	1.5164	-
0.0946	9460	1.5252	-
0.0947	9470	1.5327	-
0.0948	9480	1.5519	-
0.0949	9490	1.5077	-
0.095	9500	1.5322	0.5550
0.0951	9510	1.5358	-
0.0952	9520	1.5362	-
0.0953	9530	1.5262	-
0.0954	9540	1.5286	-
0.0955	9550	1.5205	-
0.0956	9560	1.5372	-
0.0957	9570	1.5248	-
0.0958	9580	1.5457	-
0.0959	9590	1.5087	-
0.096	9600	1.531	0.5523
0.0961	9610	1.5057	-
0.0962	9620	1.5295	-
0.0963	9630	1.52	-
0.0964	9640	1.5131	-
0.0965	9650	1.5272	-
0.0966	9660	1.5161	-
0.0967	9670	1.5178	-
0.0968	9680	1.5452	-
0.0969	9690	1.5216	-
0.097	9700	1.5471	0.5541
0.0971	9710	1.5233	-
0.0972	9720	1.5388	-
0.0973	9730	1.5173	-
0.0974	9740	1.5223	-
0.0975	9750	1.5193	-
0.0976	9760	1.5143	-
0.0977	9770	1.5245	-
0.0978	9780	1.5368	-
0.0979	9790	1.5237	-
0.098	9800	1.5077	0.5545
0.0981	9810	1.5276	-
0.0982	9820	1.5117	-
0.0983	9830	1.5174	-
0.0984	9840	1.5359	-
0.0985	9850	1.5145	-
0.0986	9860	1.5355	-
0.0987	9870	1.4959	-
0.0988	9880	1.5106	-
0.0989	9890	1.5567	-
0.099	9900	1.5102	0.5508
0.0991	9910	1.5255	-
0.0992	9920	1.4878	-
0.0993	9930	1.522	-
0.0994	9940	1.5296	-
0.0995	9950	1.4935	-
0.0996	9960	1.5081	-
0.0997	9970	1.5163	-
0.0998	9980	1.5267	-
0.0999	9990	1.5361	-
0.1	10000	1.5067	0.5510

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

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",
}

CachedMultipleNegativesRankingLoss

bibtex

@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}
}