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_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, 0.9014, 0.7471],
#         [0.9014, 1.0000, 0.6621],
#         [0.7471, 0.6621, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: validation_retrieval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.57
cosine_accuracy@3	0.78
cosine_accuracy@5	0.85
cosine_accuracy@10	0.91
cosine_precision@1	0.57
cosine_precision@3	0.4767
cosine_precision@5	0.412
cosine_precision@10	0.316
cosine_recall@1	0.0959
cosine_recall@3	0.1974
cosine_recall@5	0.2626
cosine_recall@10	0.3592
cosine_ndcg@10	0.4669
cosine_ndcg@100	0.5134
cosine_mrr@10	0.6837
cosine_mrr@100	0.6877
cosine_map@100	0.3189

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	5.2241	-
0.0001	10	3.8307	-
0.0002	20	2.8362	-
0.0003	30	2.5535	-
0.0004	40	2.3874	-
0.0005	50	2.3478	-
0.0006	60	2.2947	-
0.0007	70	2.2224	-
0.0008	80	2.1957	-
0.0009	90	2.1906	-
0.001	100	2.1638	0.4390
0.0011	110	2.141	-
0.0012	120	2.1424	-
0.0013	130	2.1075	-
0.0014	140	2.1	-
0.0015	150	2.087	-
0.0016	160	2.0809	-
0.0017	170	2.1087	-
0.0018	180	2.0668	-
0.0019	190	2.057	-
0.002	200	2.0647	0.4512
0.0021	210	2.0259	-
0.0022	220	2.0266	-
0.0023	230	2.0173	-
0.0024	240	2.0347	-
0.0025	250	2.0364	-
0.0026	260	1.9999	-
0.0027	270	2.0162	-
0.0028	280	1.9844	-
0.0029	290	1.9897	-
0.003	300	1.9756	0.4665
0.0031	310	1.9482	-
0.0032	320	1.97	-
0.0033	330	1.9571	-
0.0034	340	1.9638	-
0.0035	350	1.9598	-
0.0036	360	1.9482	-
0.0037	370	1.9507	-
0.0038	380	1.9339	-
0.0039	390	1.9431	-
0.004	400	1.9562	0.4728
0.0041	410	1.9217	-
0.0042	420	1.9493	-
0.0043	430	1.9322	-
0.0044	440	1.9046	-
0.0045	450	1.9142	-
0.0046	460	1.9117	-
0.0047	470	1.9063	-
0.0048	480	1.8879	-
0.0049	490	1.8921	-
0.005	500	1.9033	0.4813
0.0051	510	1.8904	-
0.0052	520	1.8941	-
0.0053	530	1.8901	-
0.0054	540	1.88	-
0.0055	550	1.8974	-
0.0056	560	1.8875	-
0.0057	570	1.8962	-
0.0058	580	1.9163	-
0.0059	590	1.8463	-
0.006	600	1.8859	0.4808
0.0061	610	1.8772	-
0.0062	620	1.8664	-
0.0063	630	1.8553	-
0.0064	640	1.8674	-
0.0065	650	1.8326	-
0.0066	660	1.8411	-
0.0067	670	1.8764	-
0.0068	680	1.8544	-
0.0069	690	1.8489	-
0.007	700	1.8488	0.4785
0.0071	710	1.8558	-
0.0072	720	1.8389	-
0.0073	730	1.8367	-
0.0074	740	1.8525	-
0.0075	750	1.8368	-
0.0076	760	1.8561	-
0.0077	770	1.8255	-
0.0078	780	1.8169	-
0.0079	790	1.8036	-
0.008	800	1.8187	0.4950
0.0081	810	1.835	-
0.0082	820	1.8459	-
0.0083	830	1.7894	-
0.0084	840	1.8323	-
0.0085	850	1.8184	-
0.0086	860	1.8148	-
0.0087	870	1.8335	-
0.0088	880	1.8051	-
0.0089	890	1.796	-
0.009	900	1.8298	0.4931
0.0091	910	1.8048	-
0.0092	920	1.7841	-
0.0093	930	1.8022	-
0.0094	940	1.7822	-
0.0095	950	1.8062	-
0.0096	960	1.8197	-
0.0097	970	1.7851	-
0.0098	980	1.7836	-
0.0099	990	1.7629	-
0.01	1000	1.8104	0.4907
0.0101	1010	1.7941	-
0.0102	1020	1.8084	-
0.0103	1030	1.801	-
0.0104	1040	1.782	-
0.0105	1050	1.8032	-
0.0106	1060	1.7978	-
0.0107	1070	1.758	-
0.0108	1080	1.7695	-
0.0109	1090	1.7961	-
0.011	1100	1.7667	0.4844
0.0111	1110	1.7644	-
0.0112	1120	1.7974	-
0.0113	1130	1.7672	-
0.0114	1140	1.7671	-
0.0115	1150	1.7945	-
0.0116	1160	1.7863	-
0.0117	1170	1.7613	-
0.0118	1180	1.7739	-
0.0119	1190	1.7759	-
0.012	1200	1.7736	0.4868
0.0121	1210	1.7507	-
0.0122	1220	1.7628	-
0.0123	1230	1.7723	-
0.0124	1240	1.7536	-
0.0125	1250	1.727	-
0.0126	1260	1.7445	-
0.0127	1270	1.7529	-
0.0128	1280	1.7623	-
0.0129	1290	1.7321	-
0.013	1300	1.7748	0.4892
0.0131	1310	1.7211	-
0.0132	1320	1.7745	-
0.0133	1330	1.7305	-
0.0134	1340	1.767	-
0.0135	1350	1.7551	-
0.0136	1360	1.7329	-
0.0137	1370	1.78	-
0.0138	1380	1.7495	-
0.0139	1390	1.724	-
0.014	1400	1.7436	0.4939
0.0141	1410	1.7756	-
0.0142	1420	1.7748	-
0.0143	1430	1.7508	-
0.0144	1440	1.7301	-
0.0145	1450	1.746	-
0.0146	1460	1.7387	-
0.0147	1470	1.7368	-
0.0148	1480	1.7422	-
0.0149	1490	1.7335	-
0.015	1500	1.7129	0.4923
0.0151	1510	1.731	-
0.0152	1520	1.7307	-
0.0153	1530	1.7322	-
0.0154	1540	1.7554	-
0.0155	1550	1.7125	-
0.0156	1560	1.7327	-
0.0157	1570	1.7223	-
0.0158	1580	1.7136	-
0.0159	1590	1.7267	-
0.016	1600	1.7625	0.4914
0.0161	1610	1.7267	-
0.0162	1620	1.7038	-
0.0163	1630	1.7063	-
0.0164	1640	1.7057	-
0.0165	1650	1.7284	-
0.0166	1660	1.6998	-
0.0167	1670	1.7115	-
0.0168	1680	1.7356	-
0.0169	1690	1.6929	-
0.017	1700	1.7292	0.4985
0.0171	1710	1.7109	-
0.0172	1720	1.7494	-
0.0173	1730	1.7181	-
0.0174	1740	1.712	-
0.0175	1750	1.708	-
0.0176	1760	1.7056	-
0.0177	1770	1.7039	-
0.0178	1780	1.6837	-
0.0179	1790	1.7071	-
0.018	1800	1.7121	0.4972
0.0181	1810	1.7147	-
0.0182	1820	1.7203	-
0.0183	1830	1.7023	-
0.0184	1840	1.7278	-
0.0185	1850	1.7129	-
0.0186	1860	1.7454	-
0.0187	1870	1.7011	-
0.0188	1880	1.6996	-
0.0189	1890	1.7046	-
0.019	1900	1.6877	0.4984
0.0191	1910	1.6962	-
0.0192	1920	1.7057	-
0.0193	1930	1.6968	-
0.0194	1940	1.6994	-
0.0195	1950	1.7087	-
0.0196	1960	1.6832	-
0.0197	1970	1.686	-
0.0198	1980	1.7101	-
0.0199	1990	1.7024	-
0.02	2000	1.6875	0.4987
0.0201	2010	1.6872	-
0.0202	2020	1.6918	-
0.0203	2030	1.6735	-
0.0204	2040	1.6869	-
0.0205	2050	1.7082	-
0.0206	2060	1.6991	-
0.0207	2070	1.6973	-
0.0208	2080	1.6754	-
0.0209	2090	1.6953	-
0.021	2100	1.7065	0.4954
0.0211	2110	1.6804	-
0.0212	2120	1.6705	-
0.0213	2130	1.673	-
0.0214	2140	1.6997	-
0.0215	2150	1.6774	-
0.0216	2160	1.7124	-
0.0217	2170	1.6749	-
0.0218	2180	1.6661	-
0.0219	2190	1.6782	-
0.022	2200	1.6742	0.4922
0.0221	2210	1.7204	-
0.0222	2220	1.7081	-
0.0223	2230	1.681	-
0.0224	2240	1.6775	-
0.0225	2250	1.665	-
0.0226	2260	1.6992	-
0.0227	2270	1.6531	-
0.0228	2280	1.6656	-
0.0229	2290	1.6717	-
0.023	2300	1.6922	0.4982
0.0231	2310	1.6765	-
0.0232	2320	1.6687	-
0.0233	2330	1.6682	-
0.0234	2340	1.6734	-
0.0235	2350	1.6735	-
0.0236	2360	1.6935	-
0.0237	2370	1.6753	-
0.0238	2380	1.6571	-
0.0239	2390	1.6834	-
0.024	2400	1.6964	0.4986
0.0241	2410	1.6722	-
0.0242	2420	1.655	-
0.0243	2430	1.6599	-
0.0244	2440	1.6362	-
0.0245	2450	1.6715	-
0.0246	2460	1.6889	-
0.0247	2470	1.6778	-
0.0248	2480	1.6746	-
0.0249	2490	1.6739	-
0.025	2500	1.6478	0.4960
0.0251	2510	1.6609	-
0.0252	2520	1.6575	-
0.0253	2530	1.6393	-
0.0254	2540	1.6665	-
0.0255	2550	1.637	-
0.0256	2560	1.6294	-
0.0257	2570	1.6586	-
0.0258	2580	1.6625	-
0.0259	2590	1.6465	-
0.026	2600	1.6576	0.4993
0.0261	2610	1.6683	-
0.0262	2620	1.6447	-
0.0263	2630	1.6647	-
0.0264	2640	1.641	-
0.0265	2650	1.6544	-
0.0266	2660	1.6366	-
0.0267	2670	1.6368	-
0.0268	2680	1.668	-
0.0269	2690	1.6355	-
0.027	2700	1.6621	0.5026
0.0271	2710	1.6472	-
0.0272	2720	1.6579	-
0.0273	2730	1.6631	-
0.0274	2740	1.6627	-
0.0275	2750	1.6485	-
0.0276	2760	1.655	-
0.0277	2770	1.656	-
0.0278	2780	1.6425	-
0.0279	2790	1.6207	-
0.028	2800	1.6438	0.5061
0.0281	2810	1.6466	-
0.0282	2820	1.625	-
0.0283	2830	1.6672	-
0.0284	2840	1.6154	-
0.0285	2850	1.6581	-
0.0286	2860	1.638	-
0.0287	2870	1.6252	-
0.0288	2880	1.6468	-
0.0289	2890	1.638	-
0.029	2900	1.67	0.4955
0.0291	2910	1.6236	-
0.0292	2920	1.6583	-
0.0293	2930	1.6596	-
0.0294	2940	1.6437	-
0.0295	2950	1.6362	-
0.0296	2960	1.6505	-
0.0297	2970	1.6299	-
0.0298	2980	1.6276	-
0.0299	2990	1.6274	-
0.03	3000	1.6666	0.5002
0.0301	3010	1.6358	-
0.0302	3020	1.6166	-
0.0303	3030	1.6491	-
0.0304	3040	1.6289	-
0.0305	3050	1.6544	-
0.0306	3060	1.6237	-
0.0307	3070	1.6131	-
0.0308	3080	1.6332	-
0.0309	3090	1.6182	-
0.031	3100	1.6344	0.5085
0.0311	3110	1.6217	-
0.0312	3120	1.6532	-
0.0313	3130	1.6315	-
0.0314	3140	1.6342	-
0.0315	3150	1.6281	-
0.0316	3160	1.6277	-
0.0317	3170	1.6527	-
0.0318	3180	1.6129	-
0.0319	3190	1.6247	-
0.032	3200	1.6439	0.5018
0.0321	3210	1.6422	-
0.0322	3220	1.6442	-
0.0323	3230	1.6632	-
0.0324	3240	1.6302	-
0.0325	3250	1.6162	-
0.0326	3260	1.6381	-
0.0327	3270	1.6137	-
0.0328	3280	1.6122	-
0.0329	3290	1.6224	-
0.033	3300	1.612	0.4993
0.0331	3310	1.6095	-
0.0332	3320	1.6206	-
0.0333	3330	1.6262	-
0.0334	3340	1.6052	-
0.0335	3350	1.6187	-
0.0336	3360	1.6145	-
0.0337	3370	1.6275	-
0.0338	3380	1.6093	-
0.0339	3390	1.6284	-
0.034	3400	1.6184	0.5079
0.0341	3410	1.6359	-
0.0342	3420	1.6208	-
0.0343	3430	1.6208	-
0.0344	3440	1.6341	-
0.0345	3450	1.6171	-
0.0346	3460	1.6122	-
0.0347	3470	1.6302	-
0.0348	3480	1.6214	-
0.0349	3490	1.6299	-
0.035	3500	1.6201	0.5005
0.0351	3510	1.6033	-
0.0352	3520	1.6202	-
0.0353	3530	1.6198	-
0.0354	3540	1.6207	-
0.0355	3550	1.6111	-
0.0356	3560	1.6196	-
0.0357	3570	1.6341	-
0.0358	3580	1.6086	-
0.0359	3590	1.6021	-
0.036	3600	1.6294	0.5161
0.0361	3610	1.6117	-
0.0362	3620	1.6368	-
0.0363	3630	1.6009	-
0.0364	3640	1.5983	-
0.0365	3650	1.6248	-
0.0366	3660	1.609	-
0.0367	3670	1.5975	-
0.0368	3680	1.6043	-
0.0369	3690	1.5989	-
0.037	3700	1.6164	0.5117
0.0371	3710	1.6283	-
0.0372	3720	1.5928	-
0.0373	3730	1.6104	-
0.0374	3740	1.6264	-
0.0375	3750	1.5989	-
0.0376	3760	1.5975	-
0.0377	3770	1.6011	-
0.0378	3780	1.6054	-
0.0379	3790	1.6129	-
0.038	3800	1.616	0.5119
0.0381	3810	1.618	-
0.0382	3820	1.6236	-
0.0383	3830	1.6032	-
0.0384	3840	1.6236	-
0.0385	3850	1.6003	-
0.0386	3860	1.6025	-
0.0387	3870	1.6034	-
0.0388	3880	1.599	-
0.0389	3890	1.6065	-
0.039	3900	1.6161	0.5097
0.0391	3910	1.6093	-
0.0392	3920	1.5912	-
0.0393	3930	1.5893	-
0.0394	3940	1.602	-
0.0395	3950	1.6023	-
0.0396	3960	1.6072	-
0.0397	3970	1.599	-
0.0398	3980	1.6083	-
0.0399	3990	1.5991	-
0.04	4000	1.6085	0.5089
0.0401	4010	1.5917	-
0.0402	4020	1.5934	-
0.0403	4030	1.5862	-
0.0404	4040	1.6041	-
0.0405	4050	1.6048	-
0.0406	4060	1.6145	-
0.0407	4070	1.5817	-
0.0408	4080	1.5848	-
0.0409	4090	1.6079	-
0.041	4100	1.5973	0.5101
0.0411	4110	1.6045	-
0.0412	4120	1.6083	-
0.0413	4130	1.5871	-
0.0414	4140	1.5891	-
0.0415	4150	1.5788	-
0.0416	4160	1.5859	-
0.0417	4170	1.6094	-
0.0418	4180	1.5684	-
0.0419	4190	1.5735	-
0.042	4200	1.5902	0.5126
0.0421	4210	1.578	-
0.0422	4220	1.5966	-
0.0423	4230	1.6026	-
0.0424	4240	1.6029	-
0.0425	4250	1.6013	-
0.0426	4260	1.5849	-
0.0427	4270	1.6059	-
0.0428	4280	1.6116	-
0.0429	4290	1.5928	-
0.043	4300	1.5911	0.5132
0.0431	4310	1.5873	-
0.0432	4320	1.611	-
0.0433	4330	1.5959	-
0.0434	4340	1.61	-
0.0435	4350	1.5948	-
0.0436	4360	1.5947	-
0.0437	4370	1.5969	-
0.0438	4380	1.5894	-
0.0439	4390	1.5905	-
0.044	4400	1.5826	0.5057
0.0441	4410	1.5852	-
0.0442	4420	1.6171	-
0.0443	4430	1.5904	-
0.0444	4440	1.5873	-
0.0445	4450	1.5963	-
0.0446	4460	1.6266	-
0.0447	4470	1.559	-
0.0448	4480	1.6168	-
0.0449	4490	1.599	-
0.045	4500	1.5974	0.5169
0.0451	4510	1.5921	-
0.0452	4520	1.5812	-
0.0453	4530	1.6036	-
0.0454	4540	1.5925	-
0.0455	4550	1.5967	-
0.0456	4560	1.5747	-
0.0457	4570	1.6012	-
0.0458	4580	1.6041	-
0.0459	4590	1.5917	-
0.046	4600	1.5615	0.5125
0.0461	4610	1.5794	-
0.0462	4620	1.5635	-
0.0463	4630	1.5968	-
0.0464	4640	1.6121	-
0.0465	4650	1.595	-
0.0466	4660	1.5875	-
0.0467	4670	1.5933	-
0.0468	4680	1.6045	-
0.0469	4690	1.5688	-
0.047	4700	1.5843	0.5133
0.0471	4710	1.5679	-
0.0472	4720	1.5736	-
0.0473	4730	1.5801	-
0.0474	4740	1.5753	-
0.0475	4750	1.5563	-
0.0476	4760	1.5565	-
0.0477	4770	1.5823	-
0.0478	4780	1.5866	-
0.0479	4790	1.5576	-
0.048	4800	1.5866	0.5147
0.0481	4810	1.5836	-
0.0482	4820	1.5993	-
0.0483	4830	1.5777	-
0.0484	4840	1.5642	-
0.0485	4850	1.5727	-
0.0486	4860	1.5621	-
0.0487	4870	1.5808	-
0.0488	4880	1.5604	-
0.0489	4890	1.5805	-
0.049	4900	1.5658	0.5227
0.0491	4910	1.5873	-
0.0492	4920	1.5984	-
0.0493	4930	1.5765	-
0.0494	4940	1.5659	-
0.0495	4950	1.5616	-
0.0496	4960	1.5851	-
0.0497	4970	1.5687	-
0.0498	4980	1.5758	-
0.0499	4990	1.588	-
0.05	5000	1.5564	0.5119
0.0501	5010	1.5673	-
0.0502	5020	1.5774	-
0.0503	5030	1.5763	-
0.0504	5040	1.572	-
0.0505	5050	1.5716	-
0.0506	5060	1.5634	-
0.0507	5070	1.5205	-
0.0508	5080	1.556	-
0.0509	5090	1.5651	-
0.051	5100	1.5684	0.5085
0.0511	5110	1.5629	-
0.0512	5120	1.5434	-
0.0513	5130	1.5796	-
0.0514	5140	1.574	-
0.0515	5150	1.5571	-
0.0516	5160	1.5433	-
0.0517	5170	1.5617	-
0.0518	5180	1.6022	-
0.0519	5190	1.5453	-
0.052	5200	1.5533	0.5091
0.0521	5210	1.5588	-
0.0522	5220	1.5756	-
0.0523	5230	1.5698	-
0.0524	5240	1.5916	-
0.0525	5250	1.5669	-
0.0526	5260	1.5595	-
0.0527	5270	1.5896	-
0.0528	5280	1.564	-
0.0529	5290	1.5819	-
0.053	5300	1.5674	0.5046
0.0531	5310	1.5725	-
0.0532	5320	1.5897	-
0.0533	5330	1.5674	-
0.0534	5340	1.5562	-
0.0535	5350	1.5582	-
0.0536	5360	1.5712	-
0.0537	5370	1.5384	-
0.0538	5380	1.5717	-
0.0539	5390	1.5793	-
0.054	5400	1.5731	0.5094
0.0541	5410	1.5597	-
0.0542	5420	1.5852	-
0.0543	5430	1.5932	-
0.0544	5440	1.5545	-
0.0545	5450	1.5954	-
0.0546	5460	1.5711	-
0.0547	5470	1.5441	-
0.0548	5480	1.5649	-
0.0549	5490	1.5662	-
0.055	5500	1.5433	0.5161
0.0551	5510	1.5714	-
0.0552	5520	1.5406	-
0.0553	5530	1.5391	-
0.0554	5540	1.5676	-
0.0555	5550	1.5568	-
0.0556	5560	1.5784	-
0.0557	5570	1.5715	-
0.0558	5580	1.5635	-
0.0559	5590	1.5659	-
0.056	5600	1.59	0.5109
0.0561	5610	1.5562	-
0.0562	5620	1.554	-
0.0563	5630	1.5524	-
0.0564	5640	1.5544	-
0.0565	5650	1.5805	-
0.0566	5660	1.5431	-
0.0567	5670	1.5627	-
0.0568	5680	1.5731	-
0.0569	5690	1.5795	-
0.057	5700	1.5558	0.5133
0.0571	5710	1.5365	-
0.0572	5720	1.5882	-
0.0573	5730	1.5812	-
0.0574	5740	1.5553	-
0.0575	5750	1.5516	-
0.0576	5760	1.5523	-
0.0577	5770	1.5522	-
0.0578	5780	1.5634	-
0.0579	5790	1.5467	-
0.058	5800	1.561	0.5116
0.0581	5810	1.5614	-
0.0582	5820	1.5595	-
0.0583	5830	1.5655	-
0.0584	5840	1.5842	-
0.0585	5850	1.556	-
0.0586	5860	1.5526	-
0.0587	5870	1.5637	-
0.0588	5880	1.5612	-
0.0589	5890	1.5815	-
0.059	5900	1.5407	0.5109
0.0591	5910	1.5594	-
0.0592	5920	1.5551	-
0.0593	5930	1.5485	-
0.0594	5940	1.5832	-
0.0595	5950	1.5729	-
0.0596	5960	1.567	-
0.0597	5970	1.5743	-
0.0598	5980	1.5456	-
0.0599	5990	1.5446	-
0.06	6000	1.5754	0.5088
0.0601	6010	1.5551	-
0.0602	6020	1.5345	-
0.0603	6030	1.5446	-
0.0604	6040	1.569	-
0.0605	6050	1.5573	-
0.0606	6060	1.5332	-
0.0607	6070	1.5851	-
0.0608	6080	1.5464	-
0.0609	6090	1.5552	-
0.061	6100	1.5519	0.5106
0.0611	6110	1.55	-
0.0612	6120	1.5674	-
0.0613	6130	1.5552	-
0.0614	6140	1.5582	-
0.0615	6150	1.5509	-
0.0616	6160	1.5531	-
0.0617	6170	1.5411	-
0.0618	6180	1.5569	-
0.0619	6190	1.5531	-
0.062	6200	1.5636	0.5095
0.0621	6210	1.5427	-
0.0622	6220	1.5796	-
0.0623	6230	1.5355	-
0.0624	6240	1.5428	-
0.0625	6250	1.5297	-
0.0626	6260	1.5397	-
0.0627	6270	1.5606	-
0.0628	6280	1.5366	-
0.0629	6290	1.5391	-
0.063	6300	1.5649	0.5146
0.0631	6310	1.5612	-
0.0632	6320	1.5357	-
0.0633	6330	1.5406	-
0.0634	6340	1.5636	-
0.0635	6350	1.5423	-
0.0636	6360	1.5265	-
0.0637	6370	1.5507	-
0.0638	6380	1.5481	-
0.0639	6390	1.5464	-
0.064	6400	1.5459	0.5177
0.0641	6410	1.5542	-
0.0642	6420	1.5515	-
0.0643	6430	1.5741	-
0.0644	6440	1.5338	-
0.0645	6450	1.5527	-
0.0646	6460	1.5624	-
0.0647	6470	1.5469	-
0.0648	6480	1.5486	-
0.0649	6490	1.5582	-
0.065	6500	1.536	0.5100
0.0651	6510	1.5565	-
0.0652	6520	1.5353	-
0.0653	6530	1.5236	-
0.0654	6540	1.54	-
0.0655	6550	1.5448	-
0.0656	6560	1.5495	-
0.0657	6570	1.5394	-
0.0658	6580	1.5737	-
0.0659	6590	1.5201	-
0.066	6600	1.5509	0.5127
0.0661	6610	1.5549	-
0.0662	6620	1.5545	-
0.0663	6630	1.5601	-
0.0664	6640	1.537	-
0.0665	6650	1.5451	-
0.0666	6660	1.5412	-
1.0000	6670	1.5383	-
1.0001	6680	1.5399	-
1.0002	6690	1.5336	-
1.0003	6700	1.5344	0.5156
1.0004	6710	1.5272	-
1.0005	6720	1.5436	-
1.0006	6730	1.5438	-
1.0007	6740	1.5238	-
1.0008	6750	1.527	-
1.0009	6760	1.5379	-
1.0010	6770	1.5227	-
1.0011	6780	1.5424	-
1.0012	6790	1.5154	-
1.0013	6800	1.5199	0.5088
1.0014	6810	1.5277	-
1.0015	6820	1.5371	-
1.0016	6830	1.5455	-
1.0017	6840	1.5624	-
1.0018	6850	1.5471	-
1.0019	6860	1.5287	-
1.0020	6870	1.5495	-
1.0021	6880	1.5094	-
1.0022	6890	1.5313	-
1.0023	6900	1.5386	0.5092
1.0024	6910	1.5389	-
1.0025	6920	1.5424	-
1.0026	6930	1.5342	-
1.0027	6940	1.5497	-
1.0028	6950	1.5228	-
1.0029	6960	1.5294	-
1.0030	6970	1.5264	-
1.0031	6980	1.508	-
1.0032	6990	1.5189	-
1.0033	7000	1.5118	0.5125
1.0034	7010	1.5243	-
1.0035	7020	1.5297	-
1.0036	7030	1.5139	-
1.0037	7040	1.5347	-
1.0038	7050	1.5236	-
1.0039	7060	1.538	-
1.0040	7070	1.5402	-
1.0041	7080	1.5136	-
1.0042	7090	1.5556	-
1.0043	7100	1.5203	0.5082
1.0044	7110	1.5284	-
1.0045	7120	1.5455	-
1.0046	7130	1.5158	-
1.0047	7140	1.5284	-
1.0048	7150	1.5064	-
1.0049	7160	1.5362	-
1.0050	7170	1.5289	-
1.0051	7180	1.5309	-
1.0052	7190	1.5189	-
1.0053	7200	1.4976	0.5141
1.0054	7210	1.5245	-
1.0055	7220	1.5435	-
1.0056	7230	1.5165	-
1.0057	7240	1.5469	-
1.0058	7250	1.5298	-
1.0059	7260	1.5141	-
1.0060	7270	1.5313	-
1.0061	7280	1.5166	-
1.0062	7290	1.5311	-
1.0063	7300	1.5188	0.5248
1.0064	7310	1.5217	-
1.0065	7320	1.4895	-
1.0066	7330	1.5336	-
1.0067	7340	1.5275	-
1.0068	7350	1.5246	-
1.0069	7360	1.5264	-
1.0070	7370	1.5325	-
1.0071	7380	1.5262	-
1.0072	7390	1.5147	-
1.0073	7400	1.515	0.5183
1.0074	7410	1.5398	-
1.0075	7420	1.5269	-
1.0076	7430	1.5146	-
1.0077	7440	1.5092	-
1.0078	7450	1.5034	-
1.0079	7460	1.5042	-
1.0080	7470	1.5185	-
1.0081	7480	1.537	-
1.0082	7490	1.521	-
1.0083	7500	1.4988	0.5193
1.0084	7510	1.5177	-
1.0085	7520	1.5335	-
1.0086	7530	1.4924	-
1.0087	7540	1.5419	-
1.0088	7550	1.4974	-
1.0089	7560	1.5227	-
1.0090	7570	1.5242	-
1.0091	7580	1.5008	-
1.0092	7590	1.4978	-
1.0093	7600	1.5043	0.5133
1.0094	7610	1.5041	-
1.0095	7620	1.5295	-
1.0096	7630	1.5221	-
1.0097	7640	1.502	-
1.0098	7650	1.4861	-
1.0099	7660	1.5108	-
1.0100	7670	1.5257	-
1.0101	7680	1.5215	-
1.0102	7690	1.5413	-
1.0103	7700	1.5071	0.5195
1.0104	7710	1.5172	-
1.0105	7720	1.531	-
1.0106	7730	1.5104	-
1.0107	7740	1.5009	-
1.0108	7750	1.5046	-
1.0109	7760	1.5339	-
1.0110	7770	1.4935	-
1.0111	7780	1.5211	-
1.0112	7790	1.5203	-
1.0113	7800	1.5066	0.5172
1.0114	7810	1.5154	-
1.0115	7820	1.5233	-
1.0116	7830	1.5256	-
1.0117	7840	1.5082	-
1.0118	7850	1.5092	-
1.0119	7860	1.5279	-
1.0120	7870	1.5133	-
1.0121	7880	1.5168	-
1.0122	7890	1.4994	-
1.0123	7900	1.5309	0.5175
1.0124	7910	1.503	-
1.0125	7920	1.4806	-
1.0126	7930	1.5077	-
1.0127	7940	1.4981	-
1.0128	7950	1.5086	-
1.0129	7960	1.5003	-
1.0130	7970	1.5208	-
1.0131	7980	1.4941	-
1.0132	7990	1.5226	-
1.0133	8000	1.5023	0.5128
1.0134	8010	1.5249	-
1.0135	8020	1.4965	-
1.0136	8030	1.5012	-
1.0137	8040	1.5477	-
1.0138	8050	1.5128	-
1.0139	8060	1.4758	-
1.0140	8070	1.5248	-
1.0141	8080	1.5497	-
1.0142	8090	1.5281	-
1.0143	8100	1.5217	0.5182
1.0144	8110	1.4934	-
1.0145	8120	1.527	-
1.0146	8130	1.522	-
1.0147	8140	1.4958	-
1.0148	8150	1.5086	-
1.0149	8160	1.5038	-
1.0150	8170	1.4971	-
1.0151	8180	1.5217	-
1.0152	8190	1.4871	-
1.0153	8200	1.525	0.5223
1.0154	8210	1.5179	-
1.0155	8220	1.4958	-
1.0156	8230	1.5119	-
1.0157	8240	1.5116	-
1.0158	8250	1.4997	-
1.0159	8260	1.5089	-
1.0160	8270	1.5377	-
1.0161	8280	1.502	-
1.0162	8290	1.486	-
1.0163	8300	1.5065	0.5142
1.0164	8310	1.5047	-
1.0165	8320	1.5028	-
1.0166	8330	1.4898	-
1.0167	8340	1.5046	-
1.0168	8350	1.5195	-
1.0169	8360	1.4848	-
1.0170	8370	1.5242	-
1.0171	8380	1.4994	-
1.0172	8390	1.5306	-
1.0173	8400	1.5079	0.5113
1.0174	8410	1.5143	-
1.0175	8420	1.495	-
1.0176	8430	1.492	-
1.0177	8440	1.5087	-
1.0178	8450	1.4765	-
1.0179	8460	1.5043	-
1.0180	8470	1.5014	-
1.0181	8480	1.5275	-
1.0182	8490	1.5101	-
1.0183	8500	1.5117	0.5129
1.0184	8510	1.5254	-
1.0185	8520	1.5261	-
1.0186	8530	1.5111	-
1.0187	8540	1.5176	-
1.0188	8550	1.4906	-
1.0189	8560	1.5139	-
1.0190	8570	1.4893	-
1.0191	8580	1.5022	-
1.0192	8590	1.5134	-
1.0193	8600	1.5087	0.5143
1.0194	8610	1.512	-
1.0195	8620	1.495	-
1.0196	8630	1.4875	-
1.0197	8640	1.5239	-
1.0198	8650	1.513	-
1.0199	8660	1.5124	-
1.0200	8670	1.4855	-
1.0201	8680	1.5028	-
1.0202	8690	1.4998	-
1.0203	8700	1.4937	0.5149
1.0204	8710	1.5103	-
1.0205	8720	1.5105	-
1.0206	8730	1.5062	-
1.0207	8740	1.5228	-
1.0208	8750	1.4866	-
1.0209	8760	1.5092	-
1.0210	8770	1.5088	-
1.0211	8780	1.488	-
1.0212	8790	1.5022	-
1.0213	8800	1.504	0.5097
1.0214	8810	1.5051	-
1.0215	8820	1.5027	-
1.0216	8830	1.5273	-
1.0217	8840	1.4884	-
1.0218	8850	1.4908	-
1.0219	8860	1.5074	-
1.0220	8870	1.516	-
1.0221	8880	1.5422	-
1.0222	8890	1.5174	-
1.0223	8900	1.5009	0.5129
1.0224	8910	1.4936	-
1.0225	8920	1.5076	-
1.0226	8930	1.5093	-
1.0227	8940	1.4848	-
1.0228	8950	1.4958	-
1.0229	8960	1.5092	-
1.0230	8970	1.5048	-
1.0231	8980	1.4895	-
1.0232	8990	1.4949	-
1.0233	9000	1.4988	0.5131
1.0234	9010	1.5205	-
1.0235	9020	1.4961	-
1.0236	9030	1.5235	-
1.0237	9040	1.4945	-
1.0238	9050	1.504	-
1.0239	9060	1.5147	-
1.0240	9070	1.5104	-
1.0241	9080	1.4981	-
1.0242	9090	1.5019	-
1.0243	9100	1.4843	0.5163
1.0244	9110	1.4691	-
1.0245	9120	1.5097	-
1.0246	9130	1.524	-
1.0247	9140	1.5055	-
1.0248	9150	1.5212	-
1.0249	9160	1.4888	-
1.0250	9170	1.4887	-
1.0251	9180	1.5051	-
1.0252	9190	1.491	-
1.0253	9200	1.492	0.5140
1.0254	9210	1.4771	-
1.0255	9220	1.4683	-
1.0256	9230	1.4824	-
1.0257	9240	1.4967	-
1.0258	9250	1.5054	-
1.0259	9260	1.4884	-
1.0260	9270	1.4975	-
1.0261	9280	1.4918	-
1.0262	9290	1.5143	-
1.0263	9300	1.4946	0.5168
1.0264	9310	1.4905	-
1.0265	9320	1.4981	-
1.0266	9330	1.4591	-
1.0267	9340	1.4933	-
1.0268	9350	1.5028	-
1.0269	9360	1.5035	-
1.0270	9370	1.4996	-
1.0271	9380	1.4939	-
1.0272	9390	1.505	-
1.0273	9400	1.496	0.5120
1.0274	9410	1.5085	-
1.0275	9420	1.5057	-
1.0276	9430	1.5065	-
1.0277	9440	1.4898	-
1.0278	9450	1.5031	-
1.0279	9460	1.4784	-
1.0280	9470	1.4906	-
1.0281	9480	1.4878	-
1.0282	9490	1.4931	-
1.0283	9500	1.5035	0.5159
1.0284	9510	1.4766	-
1.0285	9520	1.497	-
1.0286	9530	1.4965	-
1.0287	9540	1.4743	-
1.0288	9550	1.4836	-
1.0289	9560	1.5031	-
1.0290	9570	1.5287	-
1.0291	9580	1.4851	-
1.0292	9590	1.5061	-
1.0293	9600	1.5071	0.5191
1.0294	9610	1.4943	-
1.0295	9620	1.502	-
1.0296	9630	1.4953	-
1.0297	9640	1.4923	-
1.0298	9650	1.489	-
1.0299	9660	1.4968	-
1.0300	9670	1.5051	-
1.0301	9680	1.4736	-
1.0302	9690	1.4883	-
1.0303	9700	1.5062	0.5145
1.0304	9710	1.4961	-
1.0305	9720	1.5107	-
1.0306	9730	1.4852	-
1.0307	9740	1.4771	-
1.0308	9750	1.4911	-
1.0309	9760	1.4781	-
1.0310	9770	1.5064	-
1.0311	9780	1.4871	-
1.0312	9790	1.5042	-
1.0313	9800	1.4922	0.5190
1.0314	9810	1.4952	-
1.0315	9820	1.4875	-
1.0316	9830	1.4974	-
1.0317	9840	1.5013	-
1.0318	9850	1.4966	-
1.0319	9860	1.4861	-
1.0320	9870	1.5135	-
1.0321	9880	1.5038	-
1.0322	9890	1.5148	-
1.0323	9900	1.5164	0.5195
1.0324	9910	1.494	-
1.0325	9920	1.4848	-
1.0326	9930	1.5015	-
1.0327	9940	1.4868	-
1.0328	9950	1.4728	-
1.0329	9960	1.4989	-
1.0330	9970	1.4742	-
1.0331	9980	1.4743	-
1.0332	9990	1.4996	-
1.0333	10000	1.4748	0.5134

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}
}