LucaZilli/arctic-m-long-q-oai-v1

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-v1")
# 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
• num_train_epochs: 1
• max_steps: 20000
• 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: 1
• max_steps: 20000
• 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
0.0005	10	0.0654	-
0.001	20	0.0806	-
0.0015	30	0.0963	-
0.002	40	0.0865	-
0.0025	50	0.0798	-
0.003	60	0.0862	-
0.0035	70	0.084	-
0.004	80	0.0843	-
0.0045	90	0.0819	-
0.005	100	0.0861	-
0.0055	110	0.092	-
0.006	120	0.095	-
0.0065	130	0.0829	-
0.007	140	0.0759	-
0.0075	150	0.0749	0.0904
0.008	160	0.0667	-
0.0085	170	0.0845	-
0.009	180	0.0803	-
0.0095	190	0.0895	-
0.01	200	0.0805	-
0.0105	210	0.0808	-
0.011	220	0.0699	-
0.0115	230	0.0718	-
0.012	240	0.0817	-
0.0125	250	0.0687	-
0.013	260	0.0877	-
0.0135	270	0.0756	-
0.014	280	0.076	-
0.0145	290	0.0717	-
0.015	300	0.0822	0.0855
0.0155	310	0.0756	-
0.016	320	0.0697	-
0.0165	330	0.0812	-
0.017	340	0.0637	-
0.0175	350	0.0672	-
0.018	360	0.067	-
0.0185	370	0.0658	-
0.019	380	0.0694	-
0.0195	390	0.0664	-
0.02	400	0.075	-
0.0205	410	0.0662	-
0.021	420	0.0828	-
0.0215	430	0.0707	-
0.022	440	0.0748	-
0.0225	450	0.0762	0.0788
0.023	460	0.0724	-
0.0235	470	0.0621	-
0.024	480	0.0767	-
0.0245	490	0.0576	-
0.025	500	0.061	-
0.0255	510	0.0711	-
0.026	520	0.0742	-
0.0265	530	0.0737	-
0.027	540	0.074	-
0.0275	550	0.0723	-
0.028	560	0.0781	-
0.0285	570	0.0628	-
0.029	580	0.0636	-
0.0295	590	0.0658	-
0.03	600	0.0771	0.0794
0.0305	610	0.0832	-
0.031	620	0.0737	-
0.0315	630	0.0636	-
0.032	640	0.0825	-
0.0325	650	0.0822	-
0.033	660	0.0629	-
0.0335	670	0.0751	-
0.034	680	0.0823	-
0.0345	690	0.0553	-
0.035	700	0.0717	-
0.0355	710	0.0681	-
0.036	720	0.068	-
0.0365	730	0.065	-
0.037	740	0.0572	-
0.0375	750	0.0634	0.0766
0.038	760	0.0593	-
0.0385	770	0.0586	-
0.039	780	0.0677	-
0.0395	790	0.0622	-
0.04	800	0.087	-
0.0405	810	0.0888	-
0.041	820	0.0708	-
0.0415	830	0.0952	-
0.042	840	0.079	-
0.0425	850	0.0819	-
0.043	860	0.0748	-
0.0435	870	0.0633	-
0.044	880	0.0649	-
0.0445	890	0.0734	-
0.045	900	0.0514	0.0708
0.0455	910	0.0529	-
0.046	920	0.047	-
0.0465	930	0.0466	-
0.047	940	0.0436	-
0.0475	950	0.0493	-
0.048	960	0.0492	-
0.0485	970	0.0413	-
0.049	980	0.0464	-
0.0495	990	0.0425	-
0.05	1000	0.0454	-
0.0505	1010	0.0506	-
0.051	1020	0.0378	-
0.0515	1030	0.0379	-
0.052	1040	0.0462	-
0.0525	1050	0.0443	0.0685
0.053	1060	0.0419	-
0.0535	1070	0.05	-
0.054	1080	0.042	-
0.0545	1090	0.0437	-
0.055	1100	0.044	-
0.0555	1110	0.0387	-
0.056	1120	0.0484	-
0.0565	1130	0.0457	-
0.057	1140	0.0343	-
0.0575	1150	0.0431	-
0.058	1160	0.0397	-
0.0585	1170	0.0394	-
0.059	1180	0.0427	-
0.0595	1190	0.0469	-
0.06	1200	0.0392	0.0700
0.0605	1210	0.0492	-
0.061	1220	0.0398	-
0.0615	1230	0.0434	-
0.062	1240	0.042	-
0.0625	1250	0.0596	-
0.063	1260	0.0451	-
0.0635	1270	0.0455	-
0.064	1280	0.0554	-
0.0645	1290	0.0431	-
0.065	1300	0.0523	-
0.0655	1310	0.0531	-
0.066	1320	0.0493	-
0.0665	1330	0.0421	-
0.067	1340	0.0423	-
0.0675	1350	0.0704	0.0628
0.068	1360	0.0647	-
0.0685	1370	0.0478	-
0.069	1380	0.0534	-
0.0695	1390	0.0502	-
0.07	1400	0.0538	-
0.0705	1410	0.0688	-
0.071	1420	0.0632	-
0.0715	1430	0.0482	-
0.072	1440	0.0576	-
0.0725	1450	0.043	-
0.073	1460	0.0397	-
0.0735	1470	0.0553	-
0.074	1480	0.0536	-
0.0745	1490	0.045	-
0.075	1500	0.0464	0.0573
0.0755	1510	0.0481	-
0.076	1520	0.049	-
0.0765	1530	0.0554	-
0.077	1540	0.0555	-
0.0775	1550	0.048	-
0.078	1560	0.0491	-
0.0785	1570	0.046	-
0.079	1580	0.0535	-
0.0795	1590	0.0373	-
0.08	1600	0.0494	-
0.0805	1610	0.0413	-
0.081	1620	0.0519	-
0.0815	1630	0.0462	-
0.082	1640	0.0473	-
0.0825	1650	0.0382	0.0529
0.083	1660	0.0326	-
0.0835	1670	0.0522	-
0.084	1680	0.042	-
0.0845	1690	0.0478	-
0.085	1700	0.0352	-
0.0855	1710	0.0448	-
0.086	1720	0.0424	-
0.0865	1730	0.0437	-
0.087	1740	0.0458	-
0.0875	1750	0.041	-
0.088	1760	0.0365	-
0.0885	1770	0.0353	-
0.089	1780	0.0403	-
0.0895	1790	0.0352	-
0.09	1800	0.0515	0.0495
0.0905	1810	0.0302	-
0.091	1820	0.0342	-
1.0002	1830	0.0528	-
1.0007	1840	0.0661	-
1.0012	1850	0.064	-
1.0017	1860	0.0717	-
1.0022	1870	0.0507	-
1.0027	1880	0.0584	-
1.0032	1890	0.0577	-
1.0037	1900	0.0549	-
1.0042	1910	0.0577	-
1.0047	1920	0.061	-
1.0052	1930	0.0496	-
1.0057	1940	0.0607	-
1.0062	1950	0.0735	0.0511
1.0067	1960	0.0609	-
1.0072	1970	0.0703	-
1.0077	1980	0.0599	-
1.0082	1990	0.0593	-
1.0087	2000	0.0673	-
1.0092	2010	0.0627	-
1.0097	2020	0.0579	-
1.0102	2030	0.0679	-
1.0107	2040	0.063	-
1.0112	2050	0.0558	-
1.0117	2060	0.0548	-
1.0122	2070	0.0656	-
1.0127	2080	0.0606	-
1.0132	2090	0.0492	-
1.0137	2100	0.0549	0.0577
1.0142	2110	0.0631	-
1.0147	2120	0.0637	-
1.0152	2130	0.0534	-
1.0157	2140	0.0469	-
1.0162	2150	0.0581	-
1.0167	2160	0.0427	-
1.0172	2170	0.0532	-
1.0177	2180	0.0506	-
1.0182	2190	0.0516	-
1.0187	2200	0.0534	-
1.0192	2210	0.066	-
1.0197	2220	0.0484	-
1.0202	2230	0.0427	-
1.0207	2240	0.0586	-
1.0212	2250	0.0449	0.0547
1.0217	2260	0.0449	-
1.0222	2270	0.0529	-
1.0227	2280	0.0511	-
1.0232	2290	0.0469	-
1.0237	2300	0.0418	-
1.0242	2310	0.0586	-
1.0247	2320	0.0589	-
1.0252	2330	0.0551	-
1.0257	2340	0.0458	-
1.0262	2350	0.0563	-
1.0267	2360	0.0414	-
1.0272	2370	0.0532	-
1.0277	2380	0.0553	-
1.0282	2390	0.0516	-
1.0287	2400	0.0517	0.0548
1.0292	2410	0.0576	-
1.0297	2420	0.052	-
1.0302	2430	0.0501	-
1.0307	2440	0.0415	-
1.0312	2450	0.0523	-
1.0317	2460	0.0635	-
1.0322	2470	0.0601	-
1.0327	2480	0.045	-
1.0332	2490	0.0468	-
1.0337	2500	0.0381	-
1.0342	2510	0.0543	-
1.0347	2520	0.0469	-
1.0352	2530	0.0479	-
1.0357	2540	0.0489	-
1.0362	2550	0.0421	0.0591
1.0367	2560	0.0512	-
1.0372	2570	0.0378	-
1.0377	2580	0.0356	-
1.0382	2590	0.0354	-
1.0387	2600	0.053	-
1.0392	2610	0.0467	-
1.0397	2620	0.0479	-
1.0402	2630	0.0521	-
1.0407	2640	0.0418	-
1.0412	2650	0.0414	-
1.0417	2660	0.0524	-
1.0422	2670	0.0363	-
1.0427	2680	0.034	-
1.0432	2690	0.0335	-
1.0437	2700	0.0351	0.0551
1.0442	2710	0.0382	-
1.0447	2720	0.0415	-
1.0452	2730	0.0346	-
1.0457	2740	0.0401	-
1.0462	2750	0.0314	-
1.0467	2760	0.0364	-
1.0472	2770	0.0319	-
1.0477	2780	0.034	-
1.0482	2790	0.0425	-
1.0487	2800	0.0317	-
1.0492	2810	0.0265	-
1.0497	2820	0.0334	-
1.0502	2830	0.0226	-
1.0507	2840	0.0284	-
1.0512	2850	0.0368	0.0529
1.0517	2860	0.0317	-
1.0522	2870	0.027	-
1.0527	2880	0.0305	-
1.0532	2890	0.036	-
1.0537	2900	0.03	-
1.0542	2910	0.0285	-
1.0547	2920	0.0282	-
1.0552	2930	0.0327	-
1.0557	2940	0.0279	-
1.0562	2950	0.0268	-
1.0567	2960	0.0255	-
1.0572	2970	0.0241	-
1.0577	2980	0.0341	-
1.0582	2990	0.0271	-
1.0587	3000	0.0257	0.0558
1.0592	3010	0.0254	-
1.0597	3020	0.0268	-
1.0602	3030	0.0248	-
1.0607	3040	0.0318	-
1.0612	3050	0.033	-
1.0617	3060	0.0359	-
1.0622	3070	0.0312	-
1.0627	3080	0.0334	-
1.0632	3090	0.0329	-
1.0637	3100	0.0347	-
1.0642	3110	0.0399	-
1.0647	3120	0.0341	-
1.0652	3130	0.0394	-
1.0657	3140	0.0412	-
1.0662	3150	0.0441	0.0465
1.0667	3160	0.0393	-
1.0672	3170	0.0442	-
1.0677	3180	0.0309	-
1.0682	3190	0.0402	-
1.0687	3200	0.0381	-
1.0692	3210	0.0318	-
1.0697	3220	0.0374	-
1.0702	3230	0.0358	-
1.0707	3240	0.0367	-
1.0712	3250	0.038	-
1.0717	3260	0.0349	-
1.0722	3270	0.0292	-
1.0727	3280	0.042	-
1.0732	3290	0.0307	-
1.0737	3300	0.0385	0.0444
1.0742	3310	0.0337	-
1.0747	3320	0.0346	-
1.0752	3330	0.0412	-
1.0757	3340	0.0315	-
1.0762	3350	0.0316	-
1.0767	3360	0.0348	-
1.0772	3370	0.0362	-
1.0777	3380	0.0314	-
1.0782	3390	0.0394	-
1.0787	3400	0.0352	-
1.0792	3410	0.0296	-
1.0797	3420	0.0304	-
1.0802	3430	0.03	-
1.0807	3440	0.038	-
1.0812	3450	0.0297	0.0424
1.0817	3460	0.0393	-
1.0822	3470	0.0386	-
1.0827	3480	0.0309	-
1.0832	3490	0.0235	-
1.0837	3500	0.0297	-
1.0842	3510	0.0363	-
1.0847	3520	0.0208	-
1.0852	3530	0.0312	-
1.0857	3540	0.0271	-
1.0862	3550	0.0348	-
1.0867	3560	0.0343	-
1.0872	3570	0.0296	-
1.0877	3580	0.0348	-
1.0882	3590	0.0265	-
1.0887	3600	0.0316	0.0424
1.0892	3610	0.0291	-
1.0897	3620	0.0336	-
1.0902	3630	0.0267	-
1.0907	3640	0.0266	-
1.0912	3650	0.0291	-
2.0004	3660	0.0475	-
2.0009	3670	0.053	-
2.0014	3680	0.0548	-
2.0019	3690	0.0356	-
2.0024	3700	0.0429	-
2.0029	3710	0.062	-
2.0034	3720	0.037	-
2.0039	3730	0.0391	-
2.0044	3740	0.0395	-
2.0049	3750	0.047	0.0452
2.0054	3760	0.0414	-
2.0059	3770	0.043	-
2.0064	3780	0.0561	-
2.0069	3790	0.0493	-
2.0074	3800	0.0443	-
2.0079	3810	0.0442	-
2.0084	3820	0.0469	-
2.0089	3830	0.0414	-
2.0094	3840	0.0446	-
2.0099	3850	0.0443	-
2.0104	3860	0.0503	-
2.0109	3870	0.0394	-
2.0114	3880	0.0392	-
2.0119	3890	0.0402	-
2.0124	3900	0.0458	0.0518
2.0129	3910	0.0516	-
2.0134	3920	0.0364	-
2.0139	3930	0.037	-
2.0144	3940	0.0429	-
2.0149	3950	0.043	-
2.0154	3960	0.0413	-
2.0159	3970	0.041	-
2.0164	3980	0.0447	-
2.0169	3990	0.0416	-
2.0174	4000	0.0416	-
2.0179	4010	0.0373	-
2.0184	4020	0.042	-
2.0189	4030	0.0409	-
2.0194	4040	0.0454	-
2.0199	4050	0.0347	0.0562
2.0204	4060	0.0385	-
2.0209	4070	0.0388	-
2.0214	4080	0.0395	-
2.0219	4090	0.0332	-
2.0224	4100	0.0438	-
2.0229	4110	0.0468	-
2.0234	4120	0.0359	-
2.0239	4130	0.0488	-
2.0244	4140	0.0394	-
2.0249	4150	0.0349	-
2.0254	4160	0.0427	-
2.0259	4170	0.0417	-
2.0264	4180	0.0423	-
2.0269	4190	0.0375	-
2.0274	4200	0.0409	0.0547
2.0279	4210	0.036	-
2.0284	4220	0.0417	-
2.0289	4230	0.0394	-
2.0294	4240	0.0335	-
2.0299	4250	0.0451	-
2.0304	4260	0.0387	-
2.0309	4270	0.0336	-
2.0314	4280	0.0448	-
2.0319	4290	0.0396	-
2.0324	4300	0.0337	-
2.0329	4310	0.0438	-
2.0334	4320	0.0366	-
2.0339	4330	0.0396	-
2.0344	4340	0.038	-
2.0349	4350	0.0403	0.0529
2.0354	4360	0.0427	-
2.0359	4370	0.0461	-
2.0364	4380	0.0439	-
2.0369	4390	0.0328	-
2.0374	4400	0.0422	-
2.0379	4410	0.0387	-
2.0384	4420	0.0385	-
2.0389	4430	0.0414	-
2.0394	4440	0.034	-
2.0399	4450	0.0351	-
2.0404	4460	0.0364	-
2.0409	4470	0.038	-
2.0414	4480	0.0357	-
2.0419	4490	0.035	-
2.0424	4500	0.0518	0.0455
2.0429	4510	0.0399	-
2.0434	4520	0.0315	-
2.0439	4530	0.0311	-
2.0444	4540	0.0252	-
2.0449	4550	0.0291	-
2.0454	4560	0.0264	-
2.0459	4570	0.0212	-
2.0464	4580	0.0262	-
2.0469	4590	0.0248	-
2.0474	4600	0.0246	-
2.0479	4610	0.0222	-
2.0484	4620	0.0277	-
2.0489	4630	0.0177	-
2.0494	4640	0.0221	-
2.0499	4650	0.03	0.0527
2.0504	4660	0.0191	-
2.0509	4670	0.0168	-
2.0514	4680	0.0211	-
2.0519	4690	0.0237	-
2.0524	4700	0.0272	-
2.0529	4710	0.0213	-
2.0534	4720	0.0247	-
2.0539	4730	0.023	-
2.0544	4740	0.023	-
2.0549	4750	0.0233	-
2.0554	4760	0.0231	-
2.0559	4770	0.0203	-
2.0564	4780	0.0231	-
2.0569	4790	0.0194	-
2.0574	4800	0.0222	0.0464
2.0579	4810	0.0227	-
2.0584	4820	0.0256	-
2.0589	4830	0.0169	-
2.0594	4840	0.0209	-
2.0599	4850	0.0203	-
2.0604	4860	0.0264	-
2.0609	4870	0.0239	-
2.0614	4880	0.0282	-
2.0619	4890	0.0278	-
2.0624	4900	0.0175	-
2.0629	4910	0.0234	-
2.0634	4920	0.0253	-
2.0639	4930	0.0335	-
2.0644	4940	0.0261	-
2.0649	4950	0.0274	0.0434
2.0654	4960	0.0306	-
2.0659	4970	0.0253	-
2.0664	4980	0.0303	-
2.0669	4990	0.0335	-
2.0674	5000	0.0333	-
2.0679	5010	0.0321	-
2.0684	5020	0.0336	-
2.0689	5030	0.0266	-
2.0694	5040	0.0295	-
2.0699	5050	0.0319	-
2.0704	5060	0.0371	-
2.0709	5070	0.0284	-
2.0714	5080	0.0266	-
2.0719	5090	0.0259	-
2.0724	5100	0.0292	0.0432
2.0729	5110	0.0243	-
2.0734	5120	0.0283	-
2.0739	5130	0.0326	-
2.0744	5140	0.0268	-
2.0749	5150	0.0282	-
2.0754	5160	0.0225	-
2.0759	5170	0.0271	-
2.0764	5180	0.0286	-
2.0769	5190	0.0319	-
2.0774	5200	0.0317	-
2.0779	5210	0.0265	-
2.0784	5220	0.027	-
2.0789	5230	0.0287	-
2.0794	5240	0.0387	-
2.0799	5250	0.0274	0.0420
2.0804	5260	0.025	-
2.0809	5270	0.0289	-
2.0814	5280	0.0293	-
2.0819	5290	0.0224	-
2.0824	5300	0.03	-
2.0829	5310	0.0267	-
2.0834	5320	0.0299	-
2.0839	5330	0.0264	-
2.0844	5340	0.0242	-
2.0849	5350	0.0212	-
2.0854	5360	0.0262	-
2.0859	5370	0.0208	-
2.0864	5380	0.0262	-
2.0869	5390	0.025	-
2.0874	5400	0.0264	0.0431
2.0879	5410	0.0279	-
2.0884	5420	0.0245	-
2.0889	5430	0.0213	-
2.0894	5440	0.028	-
2.0899	5450	0.0231	-
2.0904	5460	0.0264	-
2.0909	5470	0.0241	-
3.0001	5480	0.0359	-
3.0006	5490	0.042	-
3.0011	5500	0.0344	-
3.0016	5510	0.0423	-
3.0021	5520	0.0318	-
3.0026	5530	0.0329	-
3.0031	5540	0.0395	-
3.0036	5550	0.0384	0.0466
3.0041	5560	0.0414	-
3.0046	5570	0.04	-
3.0051	5580	0.0418	-
3.0056	5590	0.0274	-
3.0061	5600	0.038	-
3.0066	5610	0.0415	-
3.0071	5620	0.0257	-
3.0076	5630	0.0469	-
3.0081	5640	0.0338	-
3.0086	5650	0.034	-
3.0091	5660	0.0397	-
3.0096	5670	0.0388	-
3.0101	5680	0.039	-
3.0106	5690	0.0364	-
3.0111	5700	0.0316	0.0493
3.0116	5710	0.0473	-
3.0121	5720	0.0417	-
3.0126	5730	0.0352	-
3.0131	5740	0.0386	-
3.0136	5750	0.0348	-
3.0141	5760	0.0351	-
3.0146	5770	0.0341	-
3.0151	5780	0.0329	-
3.0156	5790	0.0296	-
3.0161	5800	0.0399	-
3.0166	5810	0.032	-
3.0171	5820	0.0296	-
3.0176	5830	0.0334	-
3.0181	5840	0.0323	-
3.0186	5850	0.0322	0.0447
3.0191	5860	0.0329	-
3.0196	5870	0.034	-
3.0201	5880	0.0407	-
3.0206	5890	0.0384	-
3.0211	5900	0.033	-
3.0216	5910	0.0392	-
3.0221	5920	0.0418	-
3.0226	5930	0.0257	-
3.0231	5940	0.0342	-
3.0236	5950	0.0356	-
3.0241	5960	0.0308	-
3.0246	5970	0.0344	-
3.0251	5980	0.0388	-
3.0256	5990	0.0475	-
3.0261	6000	0.036	0.0432
3.0266	6010	0.0315	-
3.0271	6020	0.0282	-
3.0276	6030	0.0362	-
3.0281	6040	0.0348	-
3.0286	6050	0.0352	-
3.0291	6060	0.0359	-
3.0296	6070	0.0285	-
3.0301	6080	0.0374	-
3.0306	6090	0.0231	-
3.0311	6100	0.0378	-
3.0316	6110	0.0381	-
3.0321	6120	0.0327	-
3.0326	6130	0.0349	-
3.0331	6140	0.0341	-
3.0336	6150	0.0265	0.0477
3.0341	6160	0.0288	-
3.0346	6170	0.0313	-
3.0351	6180	0.0327	-
3.0356	6190	0.0346	-
3.0361	6200	0.033	-
3.0366	6210	0.0348	-
3.0371	6220	0.0348	-
3.0376	6230	0.0314	-
3.0381	6240	0.0296	-
3.0386	6250	0.0301	-
3.0391	6260	0.0344	-
3.0396	6270	0.0263	-
3.0401	6280	0.0357	-
3.0406	6290	0.0247	-
3.0411	6300	0.0277	0.0495
3.0416	6310	0.0242	-
3.0421	6320	0.0262	-
3.0426	6330	0.0257	-
3.0431	6340	0.0262	-
3.0436	6350	0.0272	-
3.0441	6360	0.0257	-
3.0446	6370	0.0258	-
3.0451	6380	0.0225	-
3.0456	6390	0.0202	-
3.0461	6400	0.0214	-
3.0466	6410	0.0248	-
3.0471	6420	0.0195	-
3.0476	6430	0.0218	-
3.0481	6440	0.024	-
3.0486	6450	0.0297	0.0439
3.0491	6460	0.0219	-
3.0496	6470	0.0256	-
3.0501	6480	0.0189	-
3.0506	6490	0.019	-
3.0511	6500	0.0235	-
3.0516	6510	0.0196	-
3.0521	6520	0.0192	-
3.0526	6530	0.0195	-
3.0531	6540	0.0189	-
3.0536	6550	0.0211	-
3.0541	6560	0.0224	-
3.0546	6570	0.0184	-
3.0551	6580	0.0196	-
3.0556	6590	0.0169	-
3.0561	6600	0.0257	0.0436
3.0566	6610	0.0164	-
3.0571	6620	0.018	-
3.0576	6630	0.0167	-
3.0581	6640	0.0194	-
3.0586	6650	0.0202	-
3.0591	6660	0.0163	-
3.0596	6670	0.0186	-
3.0601	6680	0.0193	-
3.0606	6690	0.0186	-
3.0611	6700	0.02	-
3.0616	6710	0.02	-
3.0621	6720	0.0198	-
3.0626	6730	0.0252	-
3.0631	6740	0.0183	-
3.0636	6750	0.0173	0.0454
3.0641	6760	0.0181	-
3.0646	6770	0.0246	-
3.0651	6780	0.0204	-
3.0656	6790	0.0241	-
3.0661	6800	0.0357	-
3.0666	6810	0.0268	-
3.0671	6820	0.0201	-
3.0676	6830	0.0232	-
3.0681	6840	0.0278	-
3.0686	6850	0.0232	-
3.0691	6860	0.0326	-
3.0696	6870	0.0267	-
3.0701	6880	0.029	-
3.0706	6890	0.0209	-
3.0711	6900	0.0236	0.0427
3.0716	6910	0.0228	-
3.0721	6920	0.0195	-
3.0726	6930	0.022	-
3.0731	6940	0.0304	-
3.0736	6950	0.0296	-
3.0741	6960	0.0238	-
3.0746	6970	0.0275	-
3.0751	6980	0.0234	-
3.0756	6990	0.0222	-
3.0761	7000	0.0222	-
3.0766	7010	0.0306	-
3.0771	7020	0.0271	-
3.0776	7030	0.0244	-
3.0781	7040	0.0268	-
3.0786	7050	0.0336	0.0419
3.0791	7060	0.0245	-
3.0796	7070	0.0202	-
3.0801	7080	0.022	-
3.0806	7090	0.0253	-
3.0811	7100	0.021	-
3.0816	7110	0.0233	-
3.0821	7120	0.0238	-
3.0826	7130	0.0241	-
3.0831	7140	0.0244	-
3.0836	7150	0.025	-
3.0841	7160	0.0279	-
3.0846	7170	0.0234	-
3.0851	7180	0.0235	-
3.0856	7190	0.0186	-
3.0861	7200	0.0191	0.0423
3.0866	7210	0.0269	-
3.0871	7220	0.0288	-
3.0876	7230	0.024	-
3.0881	7240	0.0217	-
3.0886	7250	0.0244	-
3.0891	7260	0.0226	-
3.0896	7270	0.0172	-
3.0901	7280	0.0211	-
3.0906	7290	0.0214	-
3.0911	7300	0.0208	-
4.0003	7310	0.033	-
4.0008	7320	0.0363	-
4.0013	7330	0.0401	-
4.0018	7340	0.0294	-
4.0023	7350	0.0387	0.0425
4.0028	7360	0.0258	-
4.0033	7370	0.0296	-
4.0038	7380	0.0432	-
4.0043	7390	0.0331	-
4.0048	7400	0.033	-
4.0053	7410	0.0274	-
4.0058	7420	0.0294	-
4.0063	7430	0.0377	-
4.0068	7440	0.0366	-
4.0073	7450	0.0239	-
4.0078	7460	0.0363	-
4.0083	7470	0.0267	-
4.0088	7480	0.0288	-
4.0093	7490	0.0354	-
4.0098	7500	0.0424	0.0452
4.0103	7510	0.0319	-
4.0108	7520	0.0375	-
4.0113	7530	0.0371	-
4.0118	7540	0.0363	-
4.0123	7550	0.0296	-
4.0128	7560	0.0363	-
4.0133	7570	0.034	-
4.0138	7580	0.0288	-
4.0143	7590	0.0341	-
4.0148	7600	0.0261	-
4.0153	7610	0.0321	-
4.0158	7620	0.0261	-
4.0163	7630	0.0364	-
4.0168	7640	0.0288	-
4.0173	7650	0.0309	0.0479
4.0178	7660	0.0306	-
4.0183	7670	0.0286	-
4.0188	7680	0.0255	-
4.0193	7690	0.0409	-
4.0198	7700	0.0363	-
4.0203	7710	0.0317	-
4.0208	7720	0.0335	-
4.0213	7730	0.0245	-
4.0218	7740	0.0316	-
4.0223	7750	0.0344	-
4.0228	7760	0.02	-
4.0233	7770	0.0318	-
4.0238	7780	0.0318	-
4.0243	7790	0.0306	-
4.0248	7800	0.0269	0.0437
4.0253	7810	0.0227	-
4.0258	7820	0.033	-
4.0263	7830	0.0278	-
4.0268	7840	0.0305	-
4.0273	7850	0.028	-
4.0278	7860	0.0278	-
4.0283	7870	0.0259	-
4.0288	7880	0.0272	-
4.0293	7890	0.0293	-
4.0298	7900	0.0344	-
4.0303	7910	0.033	-
4.0308	7920	0.0356	-
4.0313	7930	0.0266	-
4.0318	7940	0.0313	-
4.0323	7950	0.0279	0.0499
4.0328	7960	0.0278	-
4.0333	7970	0.0309	-
4.0338	7980	0.026	-
4.0343	7990	0.0315	-
4.0348	8000	0.0297	-
4.0353	8010	0.0282	-
4.0358	8020	0.0261	-
4.0363	8030	0.0284	-
4.0368	8040	0.0328	-
4.0373	8050	0.0282	-
4.0378	8060	0.0244	-
4.0383	8070	0.0301	-
4.0388	8080	0.0299	-
4.0393	8090	0.028	-
4.0398	8100	0.0323	0.0494
4.0403	8110	0.0266	-
4.0408	8120	0.0252	-
4.0413	8130	0.0229	-
4.0418	8140	0.0216	-
4.0423	8150	0.0226	-
4.0428	8160	0.0255	-
4.0433	8170	0.0273	-
4.0438	8180	0.0225	-
4.0443	8190	0.0195	-
4.0448	8200	0.019	-
4.0453	8210	0.0208	-
4.0458	8220	0.0211	-
4.0463	8230	0.022	-
4.0468	8240	0.0237	-
4.0473	8250	0.0217	0.0442

• 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",
}