sobamchan/t5-base-mrl-768-512-256-128-64

SentenceTransformer based on google-t5/t5-base

This is a sentence-transformers model finetuned from google-t5/t5-base on the all-nli 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: google-t5/t5-base
• Maximum Sequence Length: None tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • all-nli
• Language: en

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': None, 'do_lower_case': False, 'architecture': 'T5EncoderModel'})
  (1): Pooling({'word_embedding_dimension': 768, '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': False, '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("sentence_transformers_model_id")
# Run inference
sentences = [
    'A construction worker peeking out of a manhole while his coworker sits on the sidewalk smiling.',
    'A worker is looking out of a manhole.',
    'The workers are both inside the manhole.',
]
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]

Evaluation

Metrics

Semantic Similarity

• Datasets: sts-dev and sts-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	sts-dev	sts-test
pearson_cosine	0.837	0.8228
spearman_cosine	0.8415	0.8371

Training Details

Training Dataset

all-nli

• Dataset: all-nli at d482672
• Size: 557,850 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 6 tokens mean: 9.96 tokens max: 52 tokens	min: 5 tokens mean: 12.79 tokens max: 44 tokens	min: 4 tokens mean: 14.02 tokens max: 57 tokens

• Samples:

  anchor	positive	negative
  A person on a horse jumps over a broken down airplane.	A person is outdoors, on a horse.	A person is at a diner, ordering an omelette.
  Children smiling and waving at camera	There are children present	The kids are frowning
  A boy is jumping on skateboard in the middle of a red bridge.	The boy does a skateboarding trick.	The boy skates down the sidewalk.

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Evaluation Dataset

all-nli

• Dataset: all-nli at d482672
• Size: 6,584 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 5 tokens mean: 19.41 tokens max: 79 tokens	min: 4 tokens mean: 9.69 tokens max: 35 tokens	min: 4 tokens mean: 10.35 tokens max: 30 tokens

• Samples:

  anchor	positive	negative
  Two women are embracing while holding to go packages.	Two woman are holding packages.	The men are fighting outside a deli.
  Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.	Two kids in numbered jerseys wash their hands.	Two kids in jackets walk to school.
  A man selling donuts to a customer during a world exhibition event held in the city of Angeles	A man selling donuts to a customer.	A woman drinks her coffee in a small cafe.

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• num_train_epochs: 15
• warmup_ratio: 0.1

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• 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: 5e-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: 15
• max_steps: -1
• 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
• bf16: False
• 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: 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: 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': None, '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 Loss	sts-dev_spearman_cosine	sts-test_spearman_cosine
-1	-1	-	-	0.6510	-
0.0287	500	13.7054	7.7586	0.6724	-
0.0574	1000	11.7979	6.2385	0.7213	-
0.0860	1500	9.3927	4.9358	0.7593	-
0.1147	2000	7.613	4.3100	0.7712	-
0.1434	2500	6.557	3.9318	0.7783	-
0.1721	3000	6.0763	3.6271	0.7824	-
0.2008	3500	5.6509	3.3563	0.7849	-
0.2294	4000	5.2428	3.1218	0.7884	-
0.2581	4500	5.0149	2.9553	0.7919	-
0.2868	5000	4.7411	2.8019	0.7949	-
0.3155	5500	4.5017	2.6706	0.7983	-
0.3442	6000	4.288	2.5487	0.8033	-
0.3729	6500	4.2023	2.4345	0.8058	-
0.4015	7000	4.0595	2.3365	0.8075	-
0.4302	7500	3.8702	2.2540	0.8106	-
0.4589	8000	3.6251	2.1699	0.8127	-
0.4876	8500	3.7266	2.1117	0.8157	-
0.5163	9000	3.6235	2.0486	0.8161	-
0.5449	9500	3.4797	1.9871	0.8172	-
0.5736	10000	3.3095	1.9380	0.8197	-
0.6023	10500	3.3159	1.8787	0.8215	-
0.6310	11000	3.1906	1.8673	0.8226	-
0.6597	11500	3.1465	1.8114	0.8241	-
0.6883	12000	3.0597	1.7870	0.8278	-
0.7170	12500	3.027	1.7440	0.8246	-
0.7457	13000	2.9682	1.7153	0.8278	-
0.7744	13500	2.9704	1.6848	0.8234	-
0.8031	14000	2.8845	1.6477	0.8264	-
0.8318	14500	2.8022	1.6315	0.8279	-
0.8604	15000	2.7936	1.6118	0.8299	-
0.8891	15500	2.7953	1.5743	0.8323	-
0.9178	16000	2.7204	1.5756	0.8333	-
0.9465	16500	2.7123	1.5407	0.8316	-
0.9752	17000	2.5954	1.5227	0.8324	-
1.0038	17500	2.5479	1.4977	0.8344	-
1.0325	18000	2.4477	1.5045	0.8341	-
1.0612	18500	2.4471	1.4989	0.8332	-
1.0899	19000	2.4057	1.4664	0.8318	-
1.1186	19500	2.3712	1.4356	0.8313	-
1.1472	20000	2.3994	1.4368	0.8359	-
1.1759	20500	2.3075	1.4233	0.8341	-
1.2046	21000	2.3595	1.4223	0.8329	-
1.2333	21500	2.3321	1.4089	0.8353	-
1.2620	22000	2.2596	1.4086	0.8360	-
1.2907	22500	2.2511	1.4045	0.8392	-
1.3193	23000	2.2308	1.3734	0.8356	-
1.3480	23500	2.2461	1.3764	0.8364	-
1.3767	24000	2.2123	1.3812	0.8361	-
1.4054	24500	2.1176	1.3674	0.8414	-
1.4341	25000	2.1233	1.3669	0.8392	-
1.4627	25500	2.1555	1.3509	0.8378	-
1.4914	26000	2.0954	1.3471	0.8387	-
1.5201	26500	2.0441	1.3314	0.8379	-
1.5488	27000	1.9816	1.3388	0.8417	-
1.5775	27500	2.0938	1.2986	0.8410	-
1.6061	28000	2.0493	1.3159	0.8410	-
1.6348	28500	2.0017	1.3051	0.8387	-
1.6635	29000	1.9539	1.3239	0.8404	-
1.6922	29500	1.9492	1.2953	0.8423	-
1.7209	30000	2.0114	1.3136	0.8428	-
1.7496	30500	1.9353	1.2928	0.8435	-
1.7782	31000	1.9298	1.2709	0.8430	-
1.8069	31500	1.97	1.2791	0.8443	-
1.8356	32000	1.8782	1.2811	0.8424	-
1.8643	32500	1.8681	1.2436	0.8457	-
1.8930	33000	1.877	1.2562	0.8449	-
1.9216	33500	1.779	1.2503	0.8438	-
1.9503	34000	1.8633	1.2542	0.8453	-
1.9790	34500	1.8178	1.2455	0.8436	-
2.0077	35000	1.7697	1.2653	0.8446	-
2.0364	35500	1.6046	1.2587	0.8444	-
2.0650	36000	1.6431	1.2484	0.8480	-
2.0937	36500	1.6234	1.2426	0.8461	-
2.1224	37000	1.6287	1.2429	0.8482	-
2.1511	37500	1.6215	1.2468	0.8408	-
2.1798	38000	1.6565	1.2430	0.8429	-
2.2085	38500	1.5941	1.2227	0.8435	-
2.2371	39000	1.5237	1.2610	0.8412	-
2.2658	39500	1.6119	1.2515	0.8458	-
2.2945	40000	1.5083	1.2476	0.8413	-
2.3232	40500	1.5637	1.2311	0.8463	-
2.3519	41000	1.542	1.2318	0.8457	-
2.3805	41500	1.5675	1.2317	0.8444	-
2.4092	42000	1.5088	1.2334	0.8417	-
2.4379	42500	1.5145	1.2223	0.8430	-
2.4666	43000	1.5462	1.2253	0.8451	-
2.4953	43500	1.5305	1.2075	0.8440	-
2.5239	44000	1.4976	1.2181	0.8432	-
2.5526	44500	1.5038	1.2231	0.8463	-
2.5813	45000	1.531	1.2001	0.8486	-
2.6100	45500	1.4704	1.2016	0.8471	-
2.6387	46000	1.4932	1.1985	0.8463	-
2.6674	46500	1.4448	1.1928	0.8507	-
2.6960	47000	1.4798	1.1888	0.8471	-
2.7247	47500	1.476	1.1941	0.8449	-
2.7534	48000	1.4435	1.2013	0.8453	-
2.7821	48500	1.4087	1.2048	0.8443	-
2.8108	49000	1.4194	1.2043	0.8425	-
2.8394	49500	1.3939	1.1936	0.8440	-
2.8681	50000	1.4054	1.1815	0.8447	-
2.8968	50500	1.4397	1.1854	0.8435	-
2.9255	51000	1.4404	1.1609	0.8463	-
2.9542	51500	1.4051	1.1713	0.8465	-
2.9828	52000	1.3619	1.1693	0.8444	-
3.0115	52500	1.3274	1.1721	0.8461	-
3.0402	53000	1.2563	1.1841	0.8453	-
3.0689	53500	1.2658	1.1925	0.8459	-
3.0976	54000	1.2891	1.1791	0.8458	-
3.1263	54500	1.2996	1.1732	0.8424	-
3.1549	55000	1.2108	1.1987	0.8455	-
3.1836	55500	1.2212	1.2029	0.8490	-
3.2123	56000	1.2053	1.1822	0.8476	-
3.2410	56500	1.2452	1.1831	0.8472	-
3.2697	57000	1.2097	1.1640	0.8477	-
3.2983	57500	1.1934	1.1786	0.8509	-
3.3270	58000	1.2028	1.1872	0.8474	-
3.3557	58500	1.223	1.1894	0.8518	-
3.3844	59000	1.1942	1.1870	0.8513	-
3.4131	59500	1.2141	1.1943	0.8486	-
3.4417	60000	1.1649	1.1669	0.8469	-
3.4704	60500	1.18	1.1944	0.8475	-
3.4991	61000	1.1872	1.1915	0.8473	-
3.5278	61500	1.1397	1.1998	0.8462	-
3.5565	62000	1.1865	1.1862	0.8466	-
3.5852	62500	1.2027	1.1875	0.8460	-
3.6138	63000	1.1526	1.1791	0.8486	-
3.6425	63500	1.251	1.1825	0.8500	-
3.6712	64000	1.1684	1.1869	0.8463	-
3.6999	64500	1.1748	1.1939	0.8475	-
3.7286	65000	1.1585	1.1884	0.8470	-
3.7572	65500	1.1963	1.1739	0.8484	-
3.7859	66000	1.1937	1.1913	0.8467	-
3.8146	66500	1.166	1.1769	0.8475	-
3.8433	67000	1.1778	1.2029	0.8477	-
3.8720	67500	1.1213	1.1913	0.8493	-
3.9006	68000	1.1564	1.1943	0.8503	-
3.9293	68500	1.1333	1.1959	0.8457	-
3.9580	69000	1.1264	1.1895	0.8460	-
3.9867	69500	1.1383	1.1843	0.8476	-
4.0154	70000	1.0941	1.2052	0.8499	-
4.0441	70500	1.0238	1.1999	0.8479	-
4.0727	71000	0.9976	1.1865	0.8466	-
4.1014	71500	0.9958	1.2058	0.8466	-
4.1301	72000	1.0001	1.2044	0.8437	-
4.1588	72500	1.0617	1.2127	0.8425	-
4.1875	73000	1.0097	1.2046	0.8444	-
4.2161	73500	1.009	1.1841	0.8481	-
4.2448	74000	0.9992	1.2006	0.8466	-
4.2735	74500	0.9998	1.1929	0.8486	-
4.3022	75000	0.9653	1.1914	0.8480	-
4.3309	75500	0.9979	1.2030	0.8460	-
4.3595	76000	1.0385	1.2145	0.8462	-
4.3882	76500	1.0029	1.2264	0.8457	-
4.4169	77000	1.0097	1.2083	0.8458	-
4.4456	77500	0.9866	1.2104	0.8443	-
4.4743	78000	0.9902	1.1933	0.8476	-
4.5030	78500	1.0187	1.2064	0.8476	-
4.5316	79000	0.9883	1.2011	0.8457	-
4.5603	79500	0.9713	1.2051	0.8465	-
4.5890	80000	0.9886	1.2130	0.8464	-
4.6177	80500	1.0087	1.1961	0.8478	-
4.6464	81000	1.0224	1.1936	0.8485	-
4.6750	81500	0.9941	1.1944	0.8455	-
4.7037	82000	1.0294	1.1818	0.8467	-
4.7324	82500	0.9702	1.2070	0.8453	-
4.7611	83000	0.9772	1.1966	0.8440	-
4.7898	83500	0.9938	1.2063	0.8446	-
4.8184	84000	0.9935	1.1937	0.8472	-
4.8471	84500	1.0016	1.1847	0.8469	-
4.8758	85000	0.9872	1.1874	0.8464	-
4.9045	85500	0.9903	1.1877	0.8454	-
4.9332	86000	0.9963	1.1873	0.8415	-
4.9619	86500	0.9641	1.1748	0.8469	-
4.9905	87000	0.9721	1.1756	0.8479	-
5.0192	87500	0.9047	1.1986	0.8466	-
5.0479	88000	0.8849	1.2033	0.8469	-
5.0766	88500	0.8513	1.2142	0.8458	-
5.1053	89000	0.8656	1.1917	0.8438	-
5.1339	89500	0.8811	1.1839	0.8455	-
5.1626	90000	0.8834	1.1988	0.8416	-
5.1913	90500	0.9094	1.2037	0.8450	-
5.2200	91000	0.881	1.2078	0.8413	-
5.2487	91500	0.8506	1.2018	0.8395	-
5.2773	92000	0.9125	1.1737	0.8417	-
5.3060	92500	0.8409	1.1786	0.8429	-
5.3347	93000	0.89	1.1965	0.8436	-
5.3634	93500	0.8807	1.1873	0.8444	-
5.3921	94000	0.9001	1.1792	0.8477	-
5.4208	94500	0.8712	1.1922	0.8460	-
5.4494	95000	0.868	1.1855	0.8465	-
5.4781	95500	0.8599	1.1953	0.8426	-
5.5068	96000	0.8659	1.1800	0.8467	-
5.5355	96500	0.901	1.1805	0.8443	-
5.5642	97000	0.8289	1.1717	0.8449	-
5.5928	97500	0.8626	1.1812	0.8440	-
5.6215	98000	0.8387	1.1620	0.8472	-
5.6502	98500	0.8557	1.1752	0.8457	-
5.6789	99000	0.8474	1.1773	0.8434	-
5.7076	99500	0.8618	1.1762	0.8454	-
5.7362	100000	0.8882	1.1827	0.8441	-
5.7649	100500	0.822	1.1829	0.8459	-
5.7936	101000	0.8446	1.1897	0.8453	-
5.8223	101500	0.853	1.1965	0.8440	-
5.8510	102000	0.8437	1.1774	0.8440	-
5.8797	102500	0.8327	1.1791	0.8430	-
5.9083	103000	0.8639	1.1882	0.8425	-
5.9370	103500	0.8366	1.1775	0.8444	-
5.9657	104000	0.871	1.1592	0.8453	-
5.9944	104500	0.865	1.1832	0.8445	-
6.0231	105000	0.8127	1.1995	0.8452	-
6.0517	105500	0.7883	1.1890	0.8452	-
6.0804	106000	0.7288	1.1876	0.8456	-
6.1091	106500	0.7613	1.1993	0.8436	-
6.1378	107000	0.7386	1.2039	0.8465	-
6.1665	107500	0.7423	1.2102	0.8446	-
6.1951	108000	0.7775	1.2091	0.8444	-
6.2238	108500	0.7993	1.1935	0.8462	-
6.2525	109000	0.7818	1.1721	0.8480	-
6.2812	109500	0.8045	1.1804	0.8486	-
6.3099	110000	0.7896	1.1871	0.8478	-
6.3386	110500	0.7857	1.1764	0.8484	-
6.3672	111000	0.7231	1.1982	0.8470	-
6.3959	111500	0.7235	1.2115	0.8449	-
6.4246	112000	0.7556	1.1955	0.8462	-
6.4533	112500	0.7522	1.2030	0.8462	-
6.4820	113000	0.7573	1.1882	0.8470	-
6.5106	113500	0.7465	1.1933	0.8455	-
6.5393	114000	0.7676	1.1824	0.8458	-
6.5680	114500	0.7506	1.1958	0.8449	-
6.5967	115000	0.7884	1.2120	0.8446	-
6.6254	115500	0.7689	1.1832	0.8468	-
6.6540	116000	0.7725	1.1856	0.8470	-
6.6827	116500	0.7775	1.2023	0.8440	-
6.7114	117000	0.7218	1.1993	0.8447	-
6.7401	117500	0.7525	1.1952	0.8456	-
6.7688	118000	0.7714	1.1883	0.8461	-
6.7975	118500	0.7619	1.1996	0.8485	-
6.8261	119000	0.7719	1.2016	0.8458	-
6.8548	119500	0.7259	1.2038	0.8448	-
6.8835	120000	0.7315	1.2028	0.8455	-
6.9122	120500	0.7397	1.1850	0.8468	-
6.9409	121000	0.7862	1.1997	0.8446	-
6.9695	121500	0.7506	1.1941	0.8459	-
6.9982	122000	0.7272	1.1990	0.8477	-
7.0269	122500	0.7274	1.2014	0.8446	-
7.0556	123000	0.7318	1.2011	0.8450	-
7.0843	123500	0.6754	1.2200	0.8439	-
7.1129	124000	0.6767	1.2114	0.8444	-
7.1416	124500	0.7088	1.2152	0.8447	-
7.1703	125000	0.6977	1.2049	0.8442	-
7.1990	125500	0.6797	1.2232	0.8420	-
7.2277	126000	0.7212	1.2179	0.8436	-
7.2564	126500	0.6684	1.2094	0.8431	-
7.2850	127000	0.6796	1.2123	0.8419	-
7.3137	127500	0.685	1.2078	0.8424	-
7.3424	128000	0.6764	1.2067	0.8431	-
7.3711	128500	0.7198	1.2105	0.8424	-
7.3998	129000	0.6858	1.2232	0.8437	-
7.4284	129500	0.6916	1.2149	0.8458	-
7.4571	130000	0.7041	1.2160	0.8423	-
7.4858	130500	0.6931	1.2026	0.8425	-
7.5145	131000	0.6972	1.2106	0.8471	-
7.5432	131500	0.6767	1.2000	0.8473	-
7.5718	132000	0.6716	1.2099	0.8448	-
7.6005	132500	0.6833	1.2123	0.8437	-
7.6292	133000	0.6985	1.1939	0.8447	-
7.6579	133500	0.6839	1.2256	0.8435	-
7.6866	134000	0.7139	1.2038	0.8444	-
7.7153	134500	0.6885	1.2148	0.8452	-
7.7439	135000	0.6889	1.1966	0.8448	-
7.7726	135500	0.687	1.2092	0.8437	-
7.8013	136000	0.686	1.2168	0.8419	-
7.8300	136500	0.6783	1.2063	0.8443	-
7.8587	137000	0.6595	1.2156	0.8428	-
7.8873	137500	0.6661	1.2139	0.8431	-
7.9160	138000	0.6771	1.2171	0.8464	-
7.9447	138500	0.6628	1.2014	0.8448	-
7.9734	139000	0.6741	1.2073	0.8459	-
8.0021	139500	0.6697	1.2024	0.8425	-
8.0307	140000	0.605	1.2185	0.8447	-
8.0594	140500	0.636	1.2103	0.8428	-
8.0881	141000	0.6362	1.2035	0.8443	-
8.1168	141500	0.651	1.2027	0.8447	-
8.1455	142000	0.6309	1.2105	0.8447	-
8.1742	142500	0.6158	1.2106	0.8449	-
8.2028	143000	0.6271	1.2248	0.8449	-
8.2315	143500	0.6515	1.2267	0.8442	-
8.2602	144000	0.6574	1.2308	0.8439	-
8.2889	144500	0.6476	1.2315	0.8434	-
8.3176	145000	0.6342	1.2129	0.8441	-
8.3462	145500	0.6062	1.2196	0.8426	-
8.3749	146000	0.6126	1.2069	0.8432	-
8.4036	146500	0.6119	1.2061	0.8417	-
8.4323	147000	0.6237	1.2002	0.8433	-
8.4610	147500	0.6438	1.1901	0.8439	-
8.4896	148000	0.6647	1.1996	0.8449	-
8.5183	148500	0.6446	1.2100	0.8437	-
8.5470	149000	0.6431	1.1996	0.8444	-
8.5757	149500	0.6369	1.1980	0.8462	-
8.6044	150000	0.6087	1.2008	0.8473	-
8.6331	150500	0.6462	1.1958	0.8458	-
8.6617	151000	0.6356	1.2026	0.8444	-
8.6904	151500	0.6266	1.1928	0.8466	-
8.7191	152000	0.6197	1.2158	0.8454	-
8.7478	152500	0.624	1.1941	0.8461	-
8.7765	153000	0.5935	1.1846	0.8462	-
8.8051	153500	0.611	1.1844	0.8449	-
8.8338	154000	0.633	1.1973	0.8448	-
8.8625	154500	0.6102	1.2033	0.8456	-
8.8912	155000	0.6141	1.1977	0.8466	-
8.9199	155500	0.6321	1.1983	0.8468	-
8.9485	156000	0.6417	1.2106	0.8473	-
8.9772	156500	0.633	1.2050	0.8478	-
9.0059	157000	0.6146	1.2139	0.8471	-
9.0346	157500	0.5796	1.2246	0.8460	-
9.0633	158000	0.5789	1.2143	0.8481	-
9.0920	158500	0.5473	1.2287	0.8447	-
9.1206	159000	0.5577	1.2086	0.8464	-
9.1493	159500	0.584	1.1974	0.8473	-
9.1780	160000	0.5855	1.2095	0.8462	-
9.2067	160500	0.6077	1.2157	0.8448	-
9.2354	161000	0.5768	1.2164	0.8445	-
9.2640	161500	0.5727	1.2092	0.8471	-
9.2927	162000	0.5764	1.2107	0.8454	-
9.3214	162500	0.5563	1.2206	0.8459	-
9.3501	163000	0.5647	1.2141	0.8457	-
9.3788	163500	0.5838	1.2184	0.8452	-
9.4074	164000	0.5802	1.2129	0.8460	-
9.4361	164500	0.5731	1.2018	0.8468	-
9.4648	165000	0.5648	1.2125	0.8470	-
9.4935	165500	0.596	1.2174	0.8460	-
9.5222	166000	0.5734	1.2217	0.8444	-
9.5509	166500	0.5799	1.2303	0.8461	-
9.5795	167000	0.5705	1.2136	0.8462	-
9.6082	167500	0.5778	1.2268	0.8440	-
9.6369	168000	0.5804	1.2191	0.8466	-
9.6656	168500	0.551	1.2403	0.8454	-
9.6943	169000	0.5605	1.2146	0.8455	-
9.7229	169500	0.5887	1.2313	0.8427	-
9.7516	170000	0.5599	1.2298	0.8428	-
9.7803	170500	0.6183	1.2214	0.8448	-
9.8090	171000	0.5849	1.2299	0.8429	-
9.8377	171500	0.5769	1.2180	0.8452	-
9.8663	172000	0.588	1.2172	0.8443	-
9.8950	172500	0.5806	1.2187	0.8458	-
9.9237	173000	0.5743	1.2181	0.8452	-
9.9524	173500	0.5799	1.2217	0.8442	-
9.9811	174000	0.5794	1.2267	0.8439	-
10.0098	174500	0.563	1.2144	0.8451	-
10.0384	175000	0.5374	1.2289	0.8456	-
10.0671	175500	0.5342	1.2313	0.8456	-
10.0958	176000	0.5476	1.2259	0.8458	-
10.1245	176500	0.5493	1.2312	0.8441	-
10.1532	177000	0.5594	1.2215	0.8454	-
10.1818	177500	0.529	1.2212	0.8444	-
10.2105	178000	0.5522	1.2428	0.8439	-
10.2392	178500	0.5258	1.2259	0.8441	-
10.2679	179000	0.5485	1.2457	0.8429	-
10.2966	179500	0.5422	1.2239	0.8448	-
10.3252	180000	0.5489	1.2220	0.8441	-
10.3539	180500	0.5648	1.2257	0.8440	-
10.3826	181000	0.548	1.2227	0.8453	-
10.4113	181500	0.5448	1.2222	0.8453	-
10.4400	182000	0.5702	1.2268	0.8441	-
10.4687	182500	0.5389	1.2345	0.8447	-
10.4973	183000	0.5197	1.2455	0.8433	-
10.5260	183500	0.5394	1.2252	0.8447	-
10.5547	184000	0.5457	1.2319	0.8435	-
10.5834	184500	0.5276	1.2284	0.8442	-
10.6121	185000	0.5505	1.2186	0.8441	-
10.6407	185500	0.5633	1.2376	0.8435	-
10.6694	186000	0.5206	1.2274	0.8435	-
10.6981	186500	0.5416	1.2286	0.8427	-
10.7268	187000	0.5348	1.2263	0.8442	-
10.7555	187500	0.5155	1.2410	0.8433	-
10.7841	188000	0.5667	1.2273	0.8419	-
10.8128	188500	0.5106	1.2316	0.8419	-
10.8415	189000	0.5758	1.2288	0.8424	-
10.8702	189500	0.5472	1.2311	0.8411	-
10.8989	190000	0.5218	1.2360	0.8424	-
10.9276	190500	0.5455	1.2326	0.8424	-
10.9562	191000	0.5695	1.2229	0.8420	-
10.9849	191500	0.5676	1.2150	0.8420	-
11.0136	192000	0.5273	1.2110	0.8427	-
11.0423	192500	0.5056	1.2262	0.8419	-
11.0710	193000	0.5169	1.2328	0.8417	-
11.0996	193500	0.4961	1.2332	0.8414	-
11.1283	194000	0.5023	1.2248	0.8423	-
11.1570	194500	0.4722	1.2368	0.8420	-
11.1857	195000	0.5292	1.2343	0.8414	-
11.2144	195500	0.5204	1.2507	0.8408	-
11.2430	196000	0.5088	1.2359	0.8411	-
11.2717	196500	0.5126	1.2394	0.8409	-
11.3004	197000	0.5177	1.2308	0.8418	-
11.3291	197500	0.5136	1.2412	0.8399	-
11.3578	198000	0.5361	1.2284	0.8412	-
11.3865	198500	0.5146	1.2319	0.8411	-
11.4151	199000	0.5425	1.2229	0.8425	-
11.4438	199500	0.5187	1.2198	0.8437	-
11.4725	200000	0.5153	1.2227	0.8421	-
11.5012	200500	0.5193	1.2299	0.8416	-
11.5299	201000	0.5176	1.2376	0.8403	-
11.5585	201500	0.5416	1.2216	0.8410	-
11.5872	202000	0.5325	1.2294	0.8399	-
11.6159	202500	0.5169	1.2298	0.8415	-
11.6446	203000	0.4998	1.2369	0.8411	-
11.6733	203500	0.5112	1.2345	0.8410	-
11.7019	204000	0.5047	1.2262	0.8415	-
11.7306	204500	0.5253	1.2218	0.8420	-
11.7593	205000	0.5236	1.2192	0.8410	-
11.7880	205500	0.5	1.2148	0.8414	-
11.8167	206000	0.4913	1.2238	0.8402	-
11.8454	206500	0.5343	1.2173	0.8408	-
11.8740	207000	0.5094	1.2316	0.8398	-
11.9027	207500	0.4865	1.2267	0.8394	-
11.9314	208000	0.5239	1.2145	0.8407	-
11.9601	208500	0.5157	1.2215	0.8411	-
11.9888	209000	0.4732	1.2233	0.8409	-
12.0174	209500	0.4794	1.2155	0.8417	-
12.0461	210000	0.4855	1.2268	0.8407	-
12.0748	210500	0.4831	1.2297	0.8404	-
12.1035	211000	0.4982	1.2319	0.8406	-
12.1322	211500	0.4749	1.2326	0.8406	-
12.1608	212000	0.4656	1.2318	0.8399	-
12.1895	212500	0.4996	1.2331	0.8407	-
12.2182	213000	0.4796	1.2302	0.8413	-
12.2469	213500	0.4717	1.2317	0.8406	-
12.2756	214000	0.5028	1.2264	0.8407	-
12.3043	214500	0.4666	1.2318	0.8404	-
12.3329	215000	0.4828	1.2168	0.8412	-
12.3616	215500	0.5181	1.2186	0.8414	-
12.3903	216000	0.4686	1.2284	0.8402	-
12.4190	216500	0.5063	1.2253	0.8406	-
12.4477	217000	0.496	1.2342	0.8398	-
12.4763	217500	0.4975	1.2233	0.8416	-
12.5050	218000	0.5064	1.2148	0.8421	-
12.5337	218500	0.5004	1.2198	0.8424	-
12.5624	219000	0.4696	1.2321	0.8407	-
12.5911	219500	0.477	1.2273	0.8413	-
12.6197	220000	0.5117	1.2308	0.8411	-
12.6484	220500	0.4656	1.2417	0.8402	-
12.6771	221000	0.5057	1.2320	0.8407	-
12.7058	221500	0.4981	1.2357	0.8410	-
12.7345	222000	0.4893	1.2335	0.8412	-
12.7632	222500	0.4716	1.2356	0.8404	-
12.7918	223000	0.5005	1.2284	0.8403	-
12.8205	223500	0.4746	1.2201	0.8412	-
12.8492	224000	0.4787	1.2265	0.8409	-
12.8779	224500	0.4631	1.2308	0.8405	-
12.9066	225000	0.4914	1.2304	0.8405	-
12.9352	225500	0.4973	1.2342	0.8401	-
12.9639	226000	0.4872	1.2269	0.8405	-
12.9926	226500	0.5194	1.2193	0.8412	-
13.0213	227000	0.4625	1.2227	0.8413	-
13.0500	227500	0.4563	1.2189	0.8422	-
13.0786	228000	0.4786	1.2285	0.8411	-
13.1073	228500	0.4892	1.2301	0.8409	-
13.1360	229000	0.4574	1.2321	0.8409	-
13.1647	229500	0.4899	1.2407	0.8407	-
13.1934	230000	0.4645	1.2298	0.8407	-
13.2221	230500	0.4826	1.2382	0.8402	-
13.2507	231000	0.4573	1.2377	0.8404	-
13.2794	231500	0.4593	1.2323	0.8404	-
13.3081	232000	0.464	1.2295	0.8400	-
13.3368	232500	0.4869	1.2381	0.8392	-
13.3655	233000	0.454	1.2313	0.8395	-
13.3941	233500	0.4728	1.2321	0.8399	-
13.4228	234000	0.4427	1.2359	0.8399	-
13.4515	234500	0.4883	1.2376	0.8402	-
13.4802	235000	0.4656	1.2302	0.8409	-
13.5089	235500	0.4749	1.2296	0.8416	-
13.5375	236000	0.4681	1.2236	0.8417	-
13.5662	236500	0.4851	1.2327	0.8409	-
13.5949	237000	0.4782	1.2282	0.8416	-
13.6236	237500	0.4694	1.2285	0.8413	-
13.6523	238000	0.481	1.2297	0.8408	-
13.6809	238500	0.4774	1.2306	0.8400	-
13.7096	239000	0.4694	1.2333	0.8404	-
13.7383	239500	0.4562	1.2344	0.8412	-
13.7670	240000	0.4607	1.2269	0.8418	-
13.7957	240500	0.4912	1.2231	0.8413	-
13.8244	241000	0.4753	1.2216	0.8414	-
13.8530	241500	0.475	1.2217	0.8418	-
13.8817	242000	0.4763	1.2276	0.8411	-
13.9104	242500	0.4877	1.2259	0.8417	-
13.9391	243000	0.4727	1.2313	0.8412	-
13.9678	243500	0.4652	1.2286	0.8416	-
13.9964	244000	0.4738	1.2312	0.8413	-
14.0251	244500	0.4353	1.2315	0.8416	-
14.0538	245000	0.4911	1.2298	0.8413	-
14.0825	245500	0.4633	1.2296	0.8411	-
14.1112	246000	0.4806	1.2272	0.8414	-
14.1398	246500	0.4561	1.2291	0.8412	-
14.1685	247000	0.4335	1.2299	0.8410	-
14.1972	247500	0.4649	1.2248	0.8414	-
14.2259	248000	0.479	1.2258	0.8418	-
14.2546	248500	0.4417	1.2235	0.8419	-
14.2833	249000	0.474	1.2257	0.8415	-
14.3119	249500	0.4711	1.2274	0.8417	-
14.3406	250000	0.4578	1.2263	0.8416	-
14.3693	250500	0.4607	1.2278	0.8412	-
14.3980	251000	0.4519	1.2306	0.8411	-
14.4267	251500	0.4876	1.2316	0.8411	-
14.4553	252000	0.455	1.2352	0.8409	-
14.4840	252500	0.4374	1.2352	0.8413	-
14.5127	253000	0.432	1.2323	0.8416	-
14.5414	253500	0.4563	1.2323	0.8415	-
14.5701	254000	0.4508	1.2337	0.8412	-
14.5987	254500	0.441	1.2351	0.8411	-
14.6274	255000	0.4861	1.2340	0.8414	-
14.6561	255500	0.4625	1.2340	0.8413	-
14.6848	256000	0.4847	1.2330	0.8412	-
14.7135	256500	0.47	1.2323	0.8411	-
14.7422	257000	0.4438	1.2322	0.8411	-
14.7708	257500	0.4604	1.2315	0.8413	-
14.7995	258000	0.4509	1.2318	0.8413	-
14.8282	258500	0.4349	1.2309	0.8414	-
14.8569	259000	0.453	1.2309	0.8415	-
14.8856	259500	0.4465	1.2316	0.8415	-
14.9142	260000	0.4532	1.2319	0.8414	-
14.9429	260500	0.4811	1.2321	0.8415	-
14.9716	261000	0.4826	1.2323	0.8415	-
-1	-1	-	-	-	0.8371

Framework Versions

• Python: 3.13.0
• Sentence Transformers: 5.1.2
• Transformers: 4.57.1
• PyTorch: 2.9.1+cu128
• Accelerate: 1.11.0
• Datasets: 4.4.1
• 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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}