LucaZilli/arctic-m-long-q-oai-v3

SentenceTransformer based on Snowflake/snowflake-arctic-embed-m-long

This is a sentence-transformers model finetuned from Snowflake/snowflake-arctic-embed-m-long on the json dataset. It maps sentences & paragraphs to a 768-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
• Base model: Snowflake/snowflake-arctic-embed-m-long
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • json

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': 8192, 'do_lower_case': False}) with Transformer model: NomicBertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

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("LucaZilli/arctic-m-long-q-oai-v3")
# 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, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

json

• Dataset: json
• Columns: sentence1, sentence2, score, and split
• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Evaluation Dataset

json

• Dataset: json
• Columns: sentence1, sentence2, score, and split
• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 12
• per_device_eval_batch_size: 12
• learning_rate: 4.000000000000001e-06
• max_steps: 9291
• warmup_ratio: 0.1
• fp16: True
• load_best_model_at_end: True

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 12
• per_device_eval_batch_size: 12
• 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: 4.000000000000001e-06
• 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
• max_steps: 9291
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• warmup_steps: 0
• log_level: passive
• 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
• use_ipex: False
• bf16: False
• fp16: True
• 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: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• past_index: -1
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: True
• 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': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch
• optim_args: None
• adafactor: False
• group_by_length: False
• length_column_name: length
• 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
• 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
• dispatch_batches: None
• split_batches: None
• include_tokens_per_second: False
• include_num_input_tokens_seen: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss
0.0011	10	0.0927	-
0.0022	20	0.099	-
0.0032	30	0.0915	-
0.0043	40	0.0956	-
0.0054	50	0.0892	-
0.0065	60	0.066	-
0.0075	70	0.1027	-
0.0086	80	0.0994	-
0.0097	90	0.0811	-
0.0108	100	0.0724	-
0.0118	110	0.0845	-
0.0129	120	0.0611	-
0.0140	130	0.0732	-
0.0151	140	0.0682	-
0.0161	150	0.0823	0.0858
0.0172	160	0.0643	-
0.0183	170	0.0826	-
0.0194	180	0.091	-
0.0204	190	0.0737	-
0.0215	200	0.0724	-
0.0226	210	0.0787	-
0.0237	220	0.0831	-
0.0248	230	0.0708	-
0.0258	240	0.0755	-
0.0269	250	0.0889	-
0.0280	260	0.0788	-
0.0291	270	0.0833	-
0.0301	280	0.0707	-
0.0312	290	0.0624	-
0.0323	300	0.0614	0.0837
0.0334	310	0.0738	-
0.0344	320	0.0616	-
0.0355	330	0.0664	-
0.0366	340	0.0706	-
0.0377	350	0.0634	-
0.0387	360	0.0822	-
0.0398	370	0.0722	-
0.0409	380	0.0687	-
0.0420	390	0.0597	-
0.0431	400	0.0594	-
0.0441	410	0.0698	-
0.0452	420	0.0665	-
0.0463	430	0.0815	-
0.0474	440	0.0655	-
0.0484	450	0.0762	0.0792
0.0495	460	0.0638	-
0.0506	470	0.0499	-
0.0517	480	0.0693	-
0.0527	490	0.0576	-
0.0538	500	0.0686	-
0.0549	510	0.0625	-
0.0560	520	0.0655	-
0.0570	530	0.0619	-
0.0581	540	0.0594	-
0.0592	550	0.0699	-
0.0603	560	0.0737	-
0.0613	570	0.0627	-
0.0624	580	0.0604	-
0.0635	590	0.0659	-
0.0646	600	0.0819	0.0706
0.0657	610	0.0683	-
0.0667	620	0.0567	-
0.0678	630	0.0655	-
0.0689	640	0.0672	-
0.0700	650	0.0586	-
0.0710	660	0.0682	-
0.0721	670	0.0587	-
0.0732	680	0.078	-
0.0743	690	0.0658	-
0.0753	700	0.0612	-
0.0764	710	0.0617	-
0.0775	720	0.0659	-
0.0786	730	0.0596	-
0.0796	740	0.0664	-
0.0807	750	0.056	0.0772
0.0818	760	0.062	-
0.0829	770	0.0706	-
0.0840	780	0.0522	-
0.0850	790	0.061	-
0.0861	800	0.0635	-
0.0872	810	0.0579	-
0.0883	820	0.0595	-
0.0893	830	0.0581	-
0.0904	840	0.0606	-
0.0915	850	0.0599	-
0.0926	860	0.0848	-
0.0936	870	0.0899	-
0.0947	880	0.0631	-
0.0958	890	0.0466	-
0.0969	900	0.0419	0.0838
0.0979	910	0.0435	-
0.0990	920	0.0481	-
0.1001	930	0.0498	-
0.1012	940	0.0391	-
0.1022	950	0.037	-
0.1033	960	0.0413	-
0.1044	970	0.0404	-
0.1055	980	0.0414	-
0.1066	990	0.0453	-
0.1076	1000	0.0347	-
0.1087	1010	0.0348	-
0.1098	1020	0.0274	-
0.1109	1030	0.04	-
0.1119	1040	0.0437	-
0.1130	1050	0.042	0.0736
0.1141	1060	0.0353	-
0.1152	1070	0.0589	-
0.1162	1080	0.061	-
0.1173	1090	0.0647	-
0.1184	1100	0.0607	-
0.1195	1110	0.0674	-
0.1205	1120	0.0539	-
0.1216	1130	0.0591	-
0.1227	1140	0.06	-
0.1238	1150	0.0479	-
0.1249	1160	0.0526	-
0.1259	1170	0.0575	-
0.1270	1180	0.0502	-
0.1281	1190	0.0491	-
0.1292	1200	0.053	0.0568
0.1302	1210	0.0448	-
0.1313	1220	0.0435	-
0.1324	1230	0.0593	-
0.1335	1240	0.0423	-
0.1345	1250	0.0536	-
0.1356	1260	0.0388	-
0.1367	1270	0.0486	-
0.1378	1280	0.041	-
0.1388	1290	0.0465	-
0.1399	1300	0.0617	-
0.1410	1310	0.0499	-
0.1421	1320	0.0494	-
0.1431	1330	0.0446	-
0.1442	1340	0.045	-
0.1453	1350	0.0453	0.0516
0.1464	1360	0.0458	-
0.1475	1370	0.0359	-
0.1485	1380	0.0348	-
0.1496	1390	0.0464	-
0.1507	1400	0.042	-
0.1518	1410	0.0397	-
0.1528	1420	0.0415	-
0.1539	1430	0.0521	-
0.1550	1440	0.0394	-
0.1561	1450	0.0463	-
0.1571	1460	0.0426	-
0.1582	1470	0.0437	-
0.1593	1480	0.0524	-
0.1604	1490	0.0443	-
0.1614	1500	0.0448	0.0508
0.1625	1510	0.0487	-
0.1636	1520	0.0341	-
0.1647	1530	0.0345	-
0.1658	1540	0.0283	-
0.1668	1550	0.0342	-
0.1679	1560	0.0452	-
0.1690	1570	0.0355	-
0.1701	1580	0.035	-
0.1711	1590	0.0394	-
0.1722	1600	0.0353	-
0.1733	1610	0.0265	-
0.1744	1620	0.0316	-
0.1754	1630	0.0357	-
0.1765	1640	0.0445	-
0.1776	1650	0.035	0.0459
0.1787	1660	0.0421	-
0.1797	1670	0.0331	-
0.1808	1680	0.0296	-
0.1819	1690	0.0393	-
0.1830	1700	0.0294	-
0.1840	1710	0.039	-
0.1851	1720	0.031	-
0.1862	1730	0.0335	-
0.1873	1740	0.0331	-
0.1884	1750	0.0299	-
0.1894	1760	0.0258	-
0.1905	1770	0.0332	-
0.1916	1780	0.0292	-
0.1927	1790	0.0252	-
0.1937	1800	0.0322	0.0485
0.1948	1810	0.0315	-
0.1959	1820	0.0284	-
0.1970	1830	0.0211	-
0.1980	1840	0.0322	-
0.1991	1850	0.0289	-
0.2002	1860	0.0299	-
0.2013	1870	0.0325	-
0.2023	1880	0.0226	-
0.2034	1890	0.0228	-
0.2045	1900	0.0254	-
0.2056	1910	0.0265	-
0.2067	1920	0.0223	-
0.2077	1930	0.0298	-
0.2088	1940	0.0257	-
0.2099	1950	0.0266	0.0507
0.2110	1960	0.0274	-
0.2120	1970	0.0236	-
0.2131	1980	0.0192	-
0.2142	1990	0.0244	-
0.2153	2000	0.0283	-
0.2163	2010	0.0226	-
0.2174	2020	0.0254	-
0.2185	2030	0.0219	-
0.2196	2040	0.0264	-
0.2206	2050	0.0238	-
0.2217	2060	0.0249	-
0.2228	2070	0.022	-
0.2239	2080	0.0222	-
0.2249	2090	0.0238	-
0.2260	2100	0.0256	0.0514
0.2271	2110	0.0279	-
0.2282	2120	0.0197	-
0.2293	2130	0.0249	-
0.2303	2140	0.0264	-
0.2314	2150	0.0226	-
0.2325	2160	0.0292	-
0.2336	2170	0.028	-
0.2346	2180	0.0199	-
0.2357	2190	0.0273	-
0.2368	2200	0.0267	-
0.2379	2210	0.0287	-
0.2389	2220	0.0221	-
0.2400	2230	0.0185	-
0.2411	2240	0.023	-
0.2422	2250	0.024	0.0519
0.2432	2260	0.0224	-
0.2443	2270	0.0249	-
0.2454	2280	0.0227	-
0.2465	2290	0.0144	-
0.2476	2300	0.021	-
0.2486	2310	0.0248	-
0.2497	2320	0.0206	-
0.2508	2330	0.0203	-
0.2519	2340	0.022	-
0.2529	2350	0.0229	-
0.2540	2360	0.0216	-
0.2551	2370	0.0304	-
0.2562	2380	0.0197	-
0.2572	2390	0.0206	-
0.2583	2400	0.025	0.0554
0.2594	2410	0.0224	-
0.2605	2420	0.0211	-
0.2615	2430	0.0173	-
0.2626	2440	0.0186	-
0.2637	2450	0.0233	-
0.2648	2460	0.0182	-
0.2658	2470	0.0215	-
0.2669	2480	0.0221	-
0.2680	2490	0.019	-
0.2691	2500	0.022	-
0.2702	2510	0.0209	-
0.2712	2520	0.0224	-
0.2723	2530	0.0208	-
0.2734	2540	0.0198	-
0.2745	2550	0.0273	0.0622
0.2755	2560	0.0238	-
0.2766	2570	0.0196	-
0.2777	2580	0.0213	-
0.2788	2590	0.0231	-
0.2798	2600	0.0236	-
0.2809	2610	0.0211	-
0.2820	2620	0.0249	-
0.2831	2630	0.0191	-
0.2841	2640	0.0177	-
0.2852	2650	0.0182	-
0.2863	2660	0.0143	-
0.2874	2670	0.019	-
0.2885	2680	0.0182	-
0.2895	2690	0.0219	-
0.2906	2700	0.0209	0.0568
0.2917	2710	0.0219	-
0.2928	2720	0.0211	-
0.2938	2730	0.0182	-
0.2949	2740	0.0177	-
0.2960	2750	0.0246	-
0.2971	2760	0.0165	-
0.2981	2770	0.0216	-
0.2992	2780	0.0189	-
0.3003	2790	0.024	-
0.3014	2800	0.0215	-
0.3024	2810	0.0244	-
0.3035	2820	0.0179	-
0.3046	2830	0.018	-
0.3057	2840	0.0212	-
0.3067	2850	0.0223	0.0577
0.3078	2860	0.0258	-
0.3089	2870	0.0171	-
0.3100	2880	0.019	-
0.3111	2890	0.0206	-
0.3121	2900	0.0178	-
0.3132	2910	0.0172	-
0.3143	2920	0.0225	-
0.3154	2930	0.0433	-
0.3164	2940	0.0482	-
0.3175	2950	0.0475	-
0.3186	2960	0.0465	-
0.3197	2970	0.048	-
0.3207	2980	0.0322	-
0.3218	2990	0.0272	-
0.3229	3000	0.0232	0.0499
0.3240	3010	0.025	-
0.3250	3020	0.0196	-
0.3261	3030	0.0192	-
0.3272	3040	0.0177	-
0.3283	3050	0.0219	-
0.3294	3060	0.0178	-
0.3304	3070	0.0152	-
0.3315	3080	0.0187	-
0.3326	3090	0.0189	-
1.0002	3100	0.0315	-
1.0013	3110	0.0717	-
1.0024	3120	0.07	-
1.0034	3130	0.0613	-
1.0045	3140	0.0771	-
1.0056	3150	0.0593	0.0542
1.0067	3160	0.0671	-
1.0077	3170	0.0655	-
1.0088	3180	0.0578	-
1.0099	3190	0.0561	-
1.0110	3200	0.0577	-
1.0121	3210	0.0641	-
1.0131	3220	0.0506	-
1.0142	3230	0.0528	-
1.0153	3240	0.0477	-
1.0164	3250	0.052	-
1.0174	3260	0.0579	-
1.0185	3270	0.054	-
1.0196	3280	0.0592	-
1.0207	3290	0.0529	-
1.0217	3300	0.0556	0.0572
1.0228	3310	0.064	-
1.0239	3320	0.0564	-
1.0250	3330	0.0597	-
1.0260	3340	0.0568	-
1.0271	3350	0.0531	-
1.0282	3360	0.0517	-
1.0293	3370	0.0515	-
1.0304	3380	0.0552	-
1.0314	3390	0.0529	-
1.0325	3400	0.0448	-
1.0336	3410	0.0485	-
1.0347	3420	0.044	-
1.0357	3430	0.0474	-
1.0368	3440	0.0536	-
1.0379	3450	0.0487	0.0620
1.0390	3460	0.0611	-
1.0400	3470	0.0505	-
1.0411	3480	0.0474	-
1.0422	3490	0.0434	-
1.0433	3500	0.0448	-
1.0443	3510	0.0451	-
1.0454	3520	0.0485	-
1.0465	3530	0.0546	-
1.0476	3540	0.0467	-
1.0486	3550	0.0465	-
1.0497	3560	0.0489	-
1.0508	3570	0.0463	-
1.0519	3580	0.0501	-
1.0530	3590	0.0413	-
1.0540	3600	0.044	0.0538
1.0551	3610	0.0519	-
1.0562	3620	0.0381	-
1.0573	3630	0.0445	-
1.0583	3640	0.0407	-
1.0594	3650	0.0483	-
1.0605	3660	0.0613	-
1.0616	3670	0.0483	-
1.0626	3680	0.0407	-
1.0637	3690	0.0519	-
1.0648	3700	0.0489	-
1.0659	3710	0.0469	-
1.0669	3720	0.047	-
1.0680	3730	0.0568	-
1.0691	3740	0.0492	-
1.0702	3750	0.0391	0.0546
1.0713	3760	0.0495	-
1.0723	3770	0.0628	-
1.0734	3780	0.0444	-
1.0745	3790	0.0443	-
1.0756	3800	0.0466	-
1.0766	3810	0.0542	-
1.0777	3820	0.0485	-
1.0788	3830	0.0529	-
1.0799	3840	0.0401	-
1.0809	3850	0.0407	-
1.0820	3860	0.0515	-
1.0831	3870	0.0497	-
1.0842	3880	0.0425	-
1.0852	3890	0.0429	-
1.0863	3900	0.0523	0.0563
1.0874	3910	0.0456	-
1.0885	3920	0.0469	-
1.0895	3930	0.0395	-
1.0906	3940	0.0495	-
1.0917	3950	0.0626	-
1.0928	3960	0.0406	-
1.0939	3970	0.0397	-
1.0949	3980	0.0269	-
1.0960	3990	0.0241	-
1.0971	4000	0.0336	-
1.0982	4010	0.0256	-
1.0992	4020	0.0317	-
1.1003	4030	0.0315	-
1.1014	4040	0.025	-
1.1025	4050	0.0222	0.0463
1.1035	4060	0.0245	-
1.1046	4070	0.0321	-
1.1057	4080	0.0256	-
1.1068	4090	0.028	-
1.1078	4100	0.0195	-
1.1089	4110	0.0207	-
1.1100	4120	0.0232	-
1.1111	4130	0.0266	-
1.1122	4140	0.0271	-
1.1132	4150	0.0309	-
1.1143	4160	0.0275	-
1.1154	4170	0.0252	-
1.1165	4180	0.0431	-
1.1175	4190	0.0397	-
1.1186	4200	0.0415	0.0479
1.1197	4210	0.0391	-
1.1208	4220	0.0385	-
1.1218	4230	0.0357	-
1.1229	4240	0.0335	-
1.1240	4250	0.0329	-
1.1251	4260	0.0349	-
1.1261	4270	0.0355	-
1.1272	4280	0.0334	-
1.1283	4290	0.0335	-
1.1294	4300	0.0277	-
1.1304	4310	0.0433	-
1.1315	4320	0.0369	-
1.1326	4330	0.0306	-
1.1337	4340	0.0393	-
1.1348	4350	0.0277	0.0457
1.1358	4360	0.0358	-
1.1369	4370	0.0256	-
1.1380	4380	0.0441	-
1.1391	4390	0.0303	-
1.1401	4400	0.0348	-
1.1412	4410	0.0345	-
1.1423	4420	0.0438	-
1.1434	4430	0.0352	-
1.1444	4440	0.0304	-
1.1455	4450	0.0334	-
1.1466	4460	0.0309	-
1.1477	4470	0.0282	-
1.1487	4480	0.025	-
1.1498	4490	0.0314	-
1.1509	4500	0.0286	0.0451
1.1520	4510	0.0401	-
1.1531	4520	0.0284	-
1.1541	4530	0.0377	-
1.1552	4540	0.0289	-
1.1563	4550	0.0381	-
1.1574	4560	0.0331	-
1.1584	4570	0.0334	-
1.1595	4580	0.0409	-
1.1606	4590	0.0323	-
1.1617	4600	0.0373	-
1.1627	4610	0.0316	-
1.1638	4620	0.026	-
1.1649	4630	0.0268	-
1.1660	4640	0.0272	-
1.1670	4650	0.03	0.0447
1.1681	4660	0.0287	-
1.1692	4670	0.0314	-
1.1703	4680	0.0236	-
1.1713	4690	0.0247	-
1.1724	4700	0.0213	-
1.1735	4710	0.0206	-
1.1746	4720	0.0218	-
1.1757	4730	0.0236	-
1.1767	4740	0.024	-
1.1778	4750	0.0203	-
1.1789	4760	0.0264	-
1.1800	4770	0.0188	-
1.1810	4780	0.0244	-
1.1821	4790	0.0179	-
1.1832	4800	0.0197	0.0449
1.1843	4810	0.0142	-
1.1853	4820	0.0202	-
1.1864	4830	0.0222	-
1.1875	4840	0.0148	-
1.1886	4850	0.0223	-
1.1896	4860	0.0173	-
1.1907	4870	0.0162	-
1.1918	4880	0.0173	-
1.1929	4890	0.0192	-
1.1940	4900	0.0187	-
1.1950	4910	0.0175	-
1.1961	4920	0.0157	-
1.1972	4930	0.0185	-
1.1983	4940	0.0207	-
1.1993	4950	0.0205	0.0488
1.2004	4960	0.0165	-
1.2015	4970	0.0198	-
1.2026	4980	0.018	-
1.2036	4990	0.0178	-
1.2047	5000	0.0153	-
1.2058	5010	0.017	-
1.2069	5020	0.0178	-
1.2079	5030	0.0181	-
1.2090	5040	0.015	-
1.2101	5050	0.0193	-
1.2112	5060	0.0213	-
1.2122	5070	0.0135	-
1.2133	5080	0.0158	-
1.2144	5090	0.0188	-
1.2155	5100	0.018	0.0507
1.2166	5110	0.0155	-
1.2176	5120	0.013	-
1.2187	5130	0.0209	-
1.2198	5140	0.0177	-
1.2209	5150	0.0184	-
1.2219	5160	0.0144	-
1.2230	5170	0.0164	-
1.2241	5180	0.0165	-
1.2252	5190	0.0168	-
1.2262	5200	0.0186	-
1.2273	5210	0.0166	-
1.2284	5220	0.0165	-
1.2295	5230	0.0165	-
1.2305	5240	0.0179	-
1.2316	5250	0.0184	0.0498
1.2327	5260	0.0148	-
1.2338	5270	0.021	-
1.2349	5280	0.0173	-
1.2359	5290	0.0199	-
1.2370	5300	0.0204	-
1.2381	5310	0.0202	-
1.2392	5320	0.0138	-
1.2402	5330	0.017	-
1.2413	5340	0.0156	-
1.2424	5350	0.0173	-
1.2435	5360	0.0163	-
1.2445	5370	0.0175	-
1.2456	5380	0.0165	-
1.2467	5390	0.0132	-
1.2478	5400	0.0179	0.0573
1.2488	5410	0.0149	-
1.2499	5420	0.0171	-
1.2510	5430	0.0188	-
1.2521	5440	0.0146	-
1.2531	5450	0.0152	-
1.2542	5460	0.0171	-
1.2553	5470	0.0198	-
1.2564	5480	0.0162	-
1.2575	5490	0.0173	-
1.2585	5500	0.019	-
1.2596	5510	0.0136	-
1.2607	5520	0.0173	-
1.2618	5530	0.015	-
1.2628	5540	0.0121	-
1.2639	5550	0.0166	0.0564
1.2650	5560	0.0156	-
1.2661	5570	0.0164	-
1.2671	5580	0.017	-
1.2682	5590	0.0134	-
1.2693	5600	0.0175	-
1.2704	5610	0.0154	-
1.2714	5620	0.0203	-
1.2725	5630	0.011	-
1.2736	5640	0.0142	-
1.2747	5650	0.0204	-
1.2758	5660	0.0167	-
1.2768	5670	0.0173	-
1.2779	5680	0.0149	-
1.2790	5690	0.0182	-
1.2801	5700	0.0172	0.0537
1.2811	5710	0.02	-
1.2822	5720	0.0168	-
1.2833	5730	0.016	-
1.2844	5740	0.0148	-
1.2854	5750	0.0145	-
1.2865	5760	0.0142	-
1.2876	5770	0.0126	-
1.2887	5780	0.0155	-
1.2897	5790	0.0161	-
1.2908	5800	0.0152	-
1.2919	5810	0.0152	-
1.2930	5820	0.0125	-
1.2940	5830	0.0187	-
1.2951	5840	0.0153	-
1.2962	5850	0.021	0.0545
1.2973	5860	0.0134	-
1.2984	5870	0.0181	-
1.2994	5880	0.0208	-
1.3005	5890	0.0134	-
1.3016	5900	0.0158	-
1.3027	5910	0.0186	-
1.3037	5920	0.0182	-
1.3048	5930	0.0149	-
1.3059	5940	0.0147	-
1.3070	5950	0.0159	-
1.3080	5960	0.0174	-
1.3091	5970	0.0162	-
1.3102	5980	0.0153	-
1.3113	5990	0.0154	-
1.3123	6000	0.0145	0.0574
1.3134	6010	0.017	-
1.3145	6020	0.0161	-
1.3156	6030	0.0328	-
1.3167	6040	0.0384	-
1.3177	6050	0.0362	-
1.3188	6060	0.0291	-
1.3199	6070	0.0332	-
1.3210	6080	0.0287	-
1.3220	6090	0.0198	-
1.3231	6100	0.0189	-
1.3242	6110	0.0216	-
1.3253	6120	0.0123	-
1.3263	6130	0.0162	-
1.3274	6140	0.0167	-
1.3285	6150	0.014	0.0491
1.3296	6160	0.0193	-
1.3306	6170	0.0127	-
1.3317	6180	0.0171	-
1.3328	6190	0.0125	-
2.0004	6200	0.0305	-
2.0015	6210	0.0519	-
2.0026	6220	0.047	-
2.0037	6230	0.0588	-
2.0047	6240	0.0486	-
2.0058	6250	0.0468	-
2.0069	6260	0.0447	-
2.0080	6270	0.0506	-
2.0090	6280	0.0509	-
2.0101	6290	0.0467	-
2.0112	6300	0.0421	0.0503
2.0123	6310	0.0427	-
2.0133	6320	0.05	-
2.0144	6330	0.0352	-
2.0155	6340	0.0381	-
2.0166	6350	0.0456	-
2.0177	6360	0.044	-
2.0187	6370	0.0488	-
2.0198	6380	0.0436	-
2.0209	6390	0.0361	-
2.0220	6400	0.0548	-
2.0230	6410	0.0551	-
2.0241	6420	0.0467	-
2.0252	6430	0.0448	-
2.0263	6440	0.0422	-
2.0273	6450	0.0406	0.0497
2.0284	6460	0.038	-
2.0295	6470	0.0517	-
2.0306	6480	0.0415	-
2.0316	6490	0.0372	-
2.0327	6500	0.0432	-
2.0338	6510	0.0425	-
2.0349	6520	0.033	-
2.0359	6530	0.0456	-
2.0370	6540	0.0334	-
2.0381	6550	0.0464	-
2.0392	6560	0.0559	-
2.0403	6570	0.0392	-
2.0413	6580	0.041	-
2.0424	6590	0.0413	-
2.0435	6600	0.0354	0.0508
2.0446	6610	0.0439	-
2.0456	6620	0.0385	-
2.0467	6630	0.0366	-
2.0478	6640	0.0381	-
2.0489	6650	0.0444	-
2.0499	6660	0.045	-
2.0510	6670	0.0364	-
2.0521	6680	0.0351	-
2.0532	6690	0.031	-
2.0542	6700	0.0454	-
2.0553	6710	0.0354	-
2.0564	6720	0.0355	-
2.0575	6730	0.0377	-
2.0586	6740	0.036	-
2.0596	6750	0.0407	0.0523
2.0607	6760	0.0555	-
2.0618	6770	0.0386	-
2.0629	6780	0.0401	-
2.0639	6790	0.0515	-
2.0650	6800	0.0379	-
2.0661	6810	0.0406	-
2.0672	6820	0.0384	-
2.0682	6830	0.0416	-
2.0693	6840	0.0368	-
2.0704	6850	0.0381	-
2.0715	6860	0.0541	-
2.0725	6870	0.0428	-
2.0736	6880	0.0452	-
2.0747	6890	0.0384	-
2.0758	6900	0.0387	0.0509
2.0768	6910	0.0398	-
2.0779	6920	0.0449	-
2.0790	6930	0.0418	-
2.0801	6940	0.0437	-
2.0812	6950	0.0349	-
2.0822	6960	0.0421	-
2.0833	6970	0.0353	-
2.0844	6980	0.0373	-
2.0855	6990	0.0407	-
2.0865	7000	0.0482	-
2.0876	7010	0.0403	-
2.0887	7020	0.0415	-
2.0898	7030	0.035	-
2.0908	7040	0.0418	-
2.0919	7050	0.0469	0.0511
2.0930	7060	0.0331	-
2.0941	7070	0.028	-
2.0951	7080	0.0279	-
2.0962	7090	0.0227	-
2.0973	7100	0.0202	-
2.0984	7110	0.0231	-
2.0995	7120	0.0286	-
2.1005	7130	0.024	-
2.1016	7140	0.0207	-
2.1027	7150	0.0194	-
2.1038	7160	0.0248	-
2.1048	7170	0.0258	-
2.1059	7180	0.0237	-
2.1070	7190	0.0197	-
2.1081	7200	0.0205	0.0456
2.1091	7210	0.0184	-
2.1102	7220	0.0178	-
2.1113	7230	0.0258	-
2.1124	7240	0.0293	-
2.1134	7250	0.0213	-
2.1145	7260	0.0218	-
2.1156	7270	0.0271	-
2.1167	7280	0.0293	-
2.1177	7290	0.036	-
2.1188	7300	0.0304	-
2.1199	7310	0.0345	-
2.1210	7320	0.0361	-
2.1221	7330	0.0263	-
2.1231	7340	0.0318	-
2.1242	7350	0.0318	0.0427
2.1253	7360	0.0324	-
2.1264	7370	0.03	-
2.1274	7380	0.0305	-
2.1285	7390	0.029	-
2.1296	7400	0.0323	-
2.1307	7410	0.0304	-
2.1317	7420	0.0246	-
2.1328	7430	0.0277	-
2.1339	7440	0.0347	-
2.1350	7450	0.0218	-
2.1360	7460	0.0309	-
2.1371	7470	0.0243	-
2.1382	7480	0.0265	-
2.1393	7490	0.0273	-
2.1404	7500	0.0366	0.0440
2.1414	7510	0.0283	-
2.1425	7520	0.0377	-
2.1436	7530	0.0277	-
2.1447	7540	0.0327	-
2.1457	7550	0.023	-
2.1468	7560	0.0286	-
2.1479	7570	0.0241	-
2.1490	7580	0.0284	-
2.1500	7590	0.0282	-
2.1511	7600	0.0249	-
2.1522	7610	0.0275	-
2.1533	7620	0.0308	-
2.1543	7630	0.0294	-
2.1554	7640	0.0263	-
2.1565	7650	0.031	0.0427
2.1576	7660	0.0264	-
2.1586	7670	0.0338	-
2.1597	7680	0.0339	-
2.1608	7690	0.0334	-
2.1619	7700	0.0314	-
2.1630	7710	0.0328	-
2.1640	7720	0.0199	-
2.1651	7730	0.0268	-
2.1662	7740	0.0258	-
2.1673	7750	0.0274	-
2.1683	7760	0.026	-
2.1694	7770	0.0248	-
2.1705	7780	0.0208	-
2.1716	7790	0.0231	-
2.1726	7800	0.02	0.0447
2.1737	7810	0.0208	-
2.1748	7820	0.0174	-
2.1759	7830	0.0178	-
2.1769	7840	0.0207	-
2.1780	7850	0.0147	-
2.1791	7860	0.0224	-
2.1802	7870	0.0209	-
2.1813	7880	0.0212	-
2.1823	7890	0.018	-
2.1834	7900	0.0177	-
2.1845	7910	0.0149	-
2.1856	7920	0.0154	-
2.1866	7930	0.0143	-
2.1877	7940	0.0183	-
2.1888	7950	0.0217	0.0463
2.1899	7960	0.0118	-
2.1909	7970	0.0155	-
2.1920	7980	0.0186	-
2.1931	7990	0.0133	-
2.1942	8000	0.0184	-
2.1952	8010	0.0132	-
2.1963	8020	0.0181	-
2.1974	8030	0.0158	-
2.1985	8040	0.0204	-
2.1995	8050	0.0166	-
2.2006	8060	0.0178	-
2.2017	8070	0.0183	-
2.2028	8080	0.0159	-
2.2039	8090	0.0146	-
2.2049	8100	0.0156	0.0482
2.2060	8110	0.0153	-
2.2071	8120	0.0137	-
2.2082	8130	0.0158	-
2.2092	8140	0.0137	-
2.2103	8150	0.0174	-
2.2114	8160	0.0143	-
2.2125	8170	0.0151	-
2.2135	8180	0.0156	-
2.2146	8190	0.0142	-
2.2157	8200	0.0134	-
2.2168	8210	0.0162	-
2.2178	8220	0.015	-
2.2189	8230	0.0172	-
2.2200	8240	0.0149	-
2.2211	8250	0.0159	0.0478
2.2222	8260	0.0157	-
2.2232	8270	0.013	-
2.2243	8280	0.0176	-
2.2254	8290	0.0118	-
2.2265	8300	0.0171	-
2.2275	8310	0.0194	-
2.2286	8320	0.0135	-
2.2297	8330	0.0168	-
2.2308	8340	0.0149	-
2.2318	8350	0.0161	-
2.2329	8360	0.0156	-
2.2340	8370	0.0147	-
2.2351	8380	0.0159	-
2.2361	8390	0.0207	-
2.2372	8400	0.0145	0.0463
2.2383	8410	0.0155	-
2.2394	8420	0.016	-
2.2404	8430	0.0132	-
2.2415	8440	0.0166	-
2.2426	8450	0.0135	-
2.2437	8460	0.0128	-
2.2448	8470	0.0161	-
2.2458	8480	0.0148	-
2.2469	8490	0.0161	-
2.2480	8500	0.0124	-
2.2491	8510	0.0137	-
2.2501	8520	0.0153	-
2.2512	8530	0.0144	-
2.2523	8540	0.0147	-
2.2534	8550	0.0154	0.0490
2.2544	8560	0.0141	-
2.2555	8570	0.0148	-
2.2566	8580	0.0159	-
2.2577	8590	0.0136	-
2.2587	8600	0.0147	-
2.2598	8610	0.0144	-
2.2609	8620	0.0155	-
2.2620	8630	0.0145	-
2.2631	8640	0.0141	-
2.2641	8650	0.014	-
2.2652	8660	0.0111	-
2.2663	8670	0.0146	-
2.2674	8680	0.0136	-
2.2684	8690	0.0158	-
2.2695	8700	0.0138	0.0483
2.2706	8710	0.0155	-
2.2717	8720	0.0117	-
2.2727	8730	0.0145	-
2.2738	8740	0.013	-
2.2749	8750	0.0141	-
2.2760	8760	0.0168	-
2.2770	8770	0.0116	-
2.2781	8780	0.0144	-
2.2792	8790	0.0159	-
2.2803	8800	0.0179	-
2.2813	8810	0.015	-
2.2824	8820	0.0175	-
2.2835	8830	0.0126	-
2.2846	8840	0.0139	-
2.2857	8850	0.0125	0.0477
2.2867	8860	0.0128	-
2.2878	8870	0.0117	-
2.2889	8880	0.0131	-
2.2900	8890	0.0167	-
2.2910	8900	0.015	-
2.2921	8910	0.0111	-
2.2932	8920	0.0145	-
2.2943	8930	0.0145	-
2.2953	8940	0.0134	-
2.2964	8950	0.0154	-
2.2975	8960	0.0158	-
2.2986	8970	0.0193	-
2.2996	8980	0.0164	-
2.3007	8990	0.014	-
2.3018	9000	0.0147	0.0483
2.3029	9010	0.0173	-
2.3040	9020	0.0171	-
2.3050	9030	0.0132	-
2.3061	9040	0.0167	-
2.3072	9050	0.0149	-
2.3083	9060	0.0178	-
2.3093	9070	0.0143	-
2.3104	9080	0.013	-
2.3115	9090	0.0175	-
2.3126	9100	0.0125	-
2.3136	9110	0.0144	-
2.3147	9120	0.0159	-
2.3158	9130	0.0266	-
2.3169	9140	0.0318	-
2.3179	9150	0.034	0.0473
2.3190	9160	0.0306	-
2.3201	9170	0.0374	-
2.3212	9180	0.0207	-
2.3222	9190	0.0161	-
2.3233	9200	0.0141	-
2.3244	9210	0.0158	-
2.3255	9220	0.0149	-
2.3266	9230	0.0126	-
2.3276	9240	0.0137	-
2.3287	9250	0.0145	-
2.3298	9260	0.0155	-
2.3309	9270	0.0111	-
2.3319	9280	0.0141	-
2.3330	9290	0.0147	-

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.14
• Sentence Transformers: 3.4.1
• Transformers: 4.49.0
• PyTorch: 2.2.2
• Accelerate: 1.4.0
• Datasets: 3.3.2
• Tokenizers: 0.21.0

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",
}