sobamchan/bert-base-uncased-no-mrl

SentenceTransformer based on google-bert/bert-base-uncased

This is a sentence-transformers model finetuned from google-bert/bert-base-uncased 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-bert/bert-base-uncased
• Maximum Sequence Length: 512 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': 512, 'do_lower_case': False, 'architecture': 'BertModel'})
  (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)
# tensor([[1.0000, 0.7580, 0.3584],
#         [0.7580, 1.0000, 0.5104],
#         [0.3584, 0.5104, 1.0000]])

Evaluation

Metrics

Semantic Similarity

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

Metric	sts-dev	sts-test
pearson_cosine	0.8164	0.7845
spearman_cosine	0.8177	0.7887

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: 7 tokens mean: 10.46 tokens max: 46 tokens	min: 6 tokens mean: 12.81 tokens max: 40 tokens	min: 5 tokens mean: 13.4 tokens max: 50 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
      ],
      "matryoshka_weights": [
          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: 6 tokens mean: 17.95 tokens max: 63 tokens	min: 4 tokens mean: 9.78 tokens max: 29 tokens	min: 5 tokens mean: 10.35 tokens max: 29 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
      ],
      "matryoshka_weights": [
          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	-	-	0.5931	-
0.0287	500	1.9739	0.7625	0.8025	-
0.0574	1000	0.9598	0.5721	0.8252	-
0.0860	1500	0.7904	0.4589	0.8343	-
0.1147	2000	0.6916	0.4032	0.8364	-
0.1434	2500	0.6144	0.3675	0.8384	-
0.1721	3000	0.5811	0.3475	0.8395	-
0.2008	3500	0.5464	0.3261	0.8391	-
0.2294	4000	0.5043	0.3115	0.8366	-
0.2581	4500	0.4965	0.2986	0.8423	-
0.2868	5000	0.4744	0.2876	0.8421	-
0.3155	5500	0.4511	0.2793	0.8424	-
0.3442	6000	0.4457	0.2725	0.8431	-
0.3729	6500	0.436	0.2693	0.8413	-
0.4015	7000	0.4297	0.2603	0.8413	-
0.4302	7500	0.4006	0.2576	0.8348	-
0.4589	8000	0.3887	0.2561	0.8403	-
0.4876	8500	0.4067	0.2574	0.8414	-
0.5163	9000	0.4055	0.2501	0.8336	-
0.5449	9500	0.3806	0.2482	0.8370	-
0.5736	10000	0.3612	0.2462	0.8457	-
0.6023	10500	0.364	0.2476	0.8441	-
0.6310	11000	0.3461	0.2463	0.8446	-
0.6597	11500	0.3554	0.2411	0.8406	-
0.6883	12000	0.3382	0.2336	0.8433	-
0.7170	12500	0.3472	0.2420	0.8409	-
0.7457	13000	0.3399	0.2410	0.8402	-
0.7744	13500	0.3341	0.2346	0.8403	-
0.8031	14000	0.3189	0.2382	0.8374	-
0.8318	14500	0.316	0.2402	0.8411	-
0.8604	15000	0.3153	0.2333	0.8351	-
0.8891	15500	0.3232	0.2375	0.8427	-
0.9178	16000	0.3078	0.2404	0.8379	-
0.9465	16500	0.3066	0.2384	0.8402	-
0.9752	17000	0.2833	0.2409	0.8479	-
1.0038	17500	0.2916	0.2465	0.8396	-
1.0325	18000	0.2452	0.2501	0.8340	-
1.0612	18500	0.251	0.2482	0.8368	-
1.0899	19000	0.2531	0.2480	0.8296	-
1.1186	19500	0.2561	0.2410	0.8329	-
1.1472	20000	0.2665	0.2501	0.8416	-
1.1759	20500	0.2607	0.2439	0.8293	-
1.2046	21000	0.2589	0.2410	0.8369	-
1.2333	21500	0.2627	0.2550	0.8383	-
1.2620	22000	0.253	0.2494	0.8457	-
1.2907	22500	0.2483	0.2541	0.8451	-
1.3193	23000	0.2501	0.2528	0.8364	-
1.3480	23500	0.2584	0.2460	0.8343	-
1.3767	24000	0.255	0.2503	0.8331	-
1.4054	24500	0.2492	0.2530	0.8327	-
1.4341	25000	0.2507	0.2508	0.8363	-
1.4627	25500	0.2553	0.2627	0.8386	-
1.4914	26000	0.2475	0.2601	0.8333	-
1.5201	26500	0.2414	0.2625	0.8366	-
1.5488	27000	0.25	0.2515	0.8355	-
1.5775	27500	0.2593	0.2507	0.8364	-
1.6061	28000	0.251	0.2658	0.8380	-
1.6348	28500	0.2538	0.2560	0.8284	-
1.6635	29000	0.25	0.2625	0.8252	-
1.6922	29500	0.2435	0.2543	0.8285	-
1.7209	30000	0.2504	0.2489	0.8350	-
1.7496	30500	0.2441	0.2450	0.8322	-
1.7782	31000	0.2375	0.2454	0.8369	-
1.8069	31500	0.2513	0.2578	0.8393	-
1.8356	32000	0.2355	0.2575	0.8392	-
1.8643	32500	0.2415	0.2389	0.8397	-
1.8930	33000	0.2328	0.2493	0.8348	-
1.9216	33500	0.2237	0.2464	0.8354	-
1.9503	34000	0.232	0.2469	0.8362	-
1.9790	34500	0.2312	0.2483	0.8328	-
2.0077	35000	0.2183	0.2558	0.8431	-
2.0364	35500	0.1581	0.2625	0.8311	-
2.0650	36000	0.1733	0.2541	0.8427	-
2.0937	36500	0.1689	0.2545	0.8386	-
2.1224	37000	0.1771	0.2569	0.8408	-
2.1511	37500	0.1758	0.2529	0.8400	-
2.1798	38000	0.1834	0.2492	0.8372	-
2.2085	38500	0.1696	0.2490	0.8336	-
2.2371	39000	0.1646	0.2541	0.8368	-
2.2658	39500	0.1718	0.2578	0.8457	-
2.2945	40000	0.1631	0.2627	0.8367	-
2.3232	40500	0.1683	0.2552	0.8346	-
2.3519	41000	0.1666	0.2572	0.8387	-
2.3805	41500	0.1754	0.2523	0.8418	-
2.4092	42000	0.1631	0.2547	0.8368	-
2.4379	42500	0.1667	0.2572	0.8425	-
2.4666	43000	0.1659	0.2575	0.8461	-
2.4953	43500	0.1644	0.2556	0.8399	-
2.5239	44000	0.1691	0.2552	0.8385	-
2.5526	44500	0.1682	0.2638	0.8377	-
2.5813	45000	0.1719	0.2515	0.8373	-
2.6100	45500	0.1597	0.2548	0.8413	-
2.6387	46000	0.1645	0.2438	0.8431	-
2.6674	46500	0.1581	0.2485	0.8425	-
2.6960	47000	0.1615	0.2524	0.8379	-
2.7247	47500	0.1657	0.2549	0.8353	-
2.7534	48000	0.162	0.2441	0.8372	-
2.7821	48500	0.1546	0.2576	0.8332	-
2.8108	49000	0.1571	0.2514	0.8322	-
2.8394	49500	0.1587	0.2537	0.8282	-
2.8681	50000	0.1562	0.2506	0.8350	-
2.8968	50500	0.161	0.2496	0.8327	-
2.9255	51000	0.1631	0.2452	0.8330	-
2.9542	51500	0.1538	0.2494	0.8297	-
2.9828	52000	0.1524	0.2432	0.8329	-
3.0115	52500	0.1386	0.2641	0.8215	-
3.0402	53000	0.1155	0.2481	0.8307	-
3.0689	53500	0.1144	0.2601	0.8264	-
3.0976	54000	0.1172	0.2639	0.8362	-
3.1263	54500	0.1181	0.2593	0.8343	-
3.1549	55000	0.1206	0.2583	0.8342	-
3.1836	55500	0.1199	0.2614	0.8275	-
3.2123	56000	0.1127	0.2559	0.8322	-
3.2410	56500	0.1192	0.2634	0.8281	-
3.2697	57000	0.115	0.2690	0.8278	-
3.2983	57500	0.1163	0.2616	0.8320	-
3.3270	58000	0.1178	0.2441	0.8375	-
3.3557	58500	0.1219	0.2650	0.8320	-
3.3844	59000	0.1184	0.2642	0.8260	-
3.4131	59500	0.1202	0.2569	0.8320	-
3.4417	60000	0.1133	0.2634	0.8287	-
3.4704	60500	0.1136	0.2610	0.8338	-
3.4991	61000	0.1134	0.2552	0.8350	-
3.5278	61500	0.1156	0.2541	0.8341	-
3.5565	62000	0.1215	0.2516	0.8312	-
3.5852	62500	0.12	0.2528	0.8292	-
3.6138	63000	0.115	0.2600	0.8314	-
3.6425	63500	0.1259	0.2573	0.8327	-
3.6712	64000	0.116	0.2604	0.8345	-
3.6999	64500	0.1139	0.2702	0.8301	-
3.7286	65000	0.1178	0.2490	0.8342	-
3.7572	65500	0.1234	0.2556	0.8264	-
3.7859	66000	0.1205	0.2519	0.8223	-
3.8146	66500	0.1155	0.2504	0.8270	-
3.8433	67000	0.1203	0.2511	0.8377	-
3.8720	67500	0.1211	0.2513	0.8339	-
3.9006	68000	0.1203	0.2476	0.8351	-
3.9293	68500	0.112	0.2585	0.8283	-
3.9580	69000	0.1129	0.2551	0.8286	-
3.9867	69500	0.1186	0.2442	0.8354	-
4.0154	70000	0.102	0.2532	0.8339	-
4.0441	70500	0.0869	0.2557	0.8264	-
4.0727	71000	0.091	0.2516	0.8357	-
4.1014	71500	0.092	0.2560	0.8325	-
4.1301	72000	0.0837	0.2599	0.8314	-
4.1588	72500	0.0976	0.2672	0.8245	-
4.1875	73000	0.0886	0.2661	0.8191	-
4.2161	73500	0.0854	0.2664	0.8257	-
4.2448	74000	0.0886	0.2659	0.8308	-
4.2735	74500	0.0889	0.2587	0.8327	-
4.3022	75000	0.0911	0.2577	0.8364	-
4.3309	75500	0.0901	0.2674	0.8315	-
4.3595	76000	0.0926	0.2626	0.8344	-
4.3882	76500	0.0946	0.2630	0.8302	-
4.4169	77000	0.0901	0.2556	0.8283	-
4.4456	77500	0.0881	0.2521	0.8325	-
4.4743	78000	0.0923	0.2630	0.8272	-
4.5030	78500	0.0912	0.2600	0.8333	-
4.5316	79000	0.0866	0.2599	0.8296	-
4.5603	79500	0.0925	0.2568	0.8284	-
4.5890	80000	0.0888	0.2586	0.8255	-
4.6177	80500	0.0934	0.2539	0.8320	-
4.6464	81000	0.0942	0.2633	0.8333	-
4.6750	81500	0.0925	0.2544	0.8372	-
4.7037	82000	0.0933	0.2652	0.8314	-
4.7324	82500	0.0884	0.2566	0.8321	-
4.7611	83000	0.0942	0.2476	0.8338	-
4.7898	83500	0.0946	0.2634	0.8327	-
4.8184	84000	0.0913	0.2588	0.8304	-
4.8471	84500	0.0988	0.2553	0.8295	-
4.8758	85000	0.0886	0.2591	0.8292	-
4.9045	85500	0.092	0.2628	0.8306	-
4.9332	86000	0.0939	0.2643	0.8282	-
4.9619	86500	0.0859	0.2685	0.8287	-
4.9905	87000	0.089	0.2620	0.8265	-
5.0192	87500	0.0784	0.2660	0.8275	-
5.0479	88000	0.0721	0.2644	0.8236	-
5.0766	88500	0.0701	0.2681	0.8311	-
5.1053	89000	0.07	0.2627	0.8294	-
5.1339	89500	0.0708	0.2654	0.8320	-
5.1626	90000	0.068	0.2618	0.8305	-
5.1913	90500	0.0739	0.2651	0.8287	-
5.2200	91000	0.0735	0.2697	0.8322	-
5.2487	91500	0.0722	0.2776	0.8254	-
5.2773	92000	0.0748	0.2656	0.8252	-
5.3060	92500	0.0694	0.2654	0.8261	-
5.3347	93000	0.0741	0.2689	0.8271	-
5.3634	93500	0.0738	0.2725	0.8219	-
5.3921	94000	0.073	0.2728	0.8281	-
5.4208	94500	0.0737	0.2710	0.8299	-
5.4494	95000	0.0733	0.2754	0.8275	-
5.4781	95500	0.076	0.2724	0.8291	-
5.5068	96000	0.0707	0.2728	0.8293	-
5.5355	96500	0.0766	0.2720	0.8290	-
5.5642	97000	0.0722	0.2688	0.8260	-
5.5928	97500	0.0741	0.2661	0.8248	-
5.6215	98000	0.0714	0.2636	0.8226	-
5.6502	98500	0.0715	0.2641	0.8292	-
5.6789	99000	0.0712	0.2698	0.8282	-
5.7076	99500	0.0771	0.2620	0.8280	-
5.7362	100000	0.0807	0.2726	0.8298	-
5.7649	100500	0.0667	0.2710	0.8312	-
5.7936	101000	0.0678	0.2714	0.8277	-
5.8223	101500	0.079	0.2729	0.8253	-
5.8510	102000	0.0732	0.2674	0.8263	-
5.8797	102500	0.0708	0.2605	0.8228	-
5.9083	103000	0.0813	0.2587	0.8279	-
5.9370	103500	0.0688	0.2633	0.8289	-
5.9657	104000	0.0787	0.2561	0.8323	-
5.9944	104500	0.0744	0.2604	0.8290	-
6.0231	105000	0.0651	0.2618	0.8299	-
6.0517	105500	0.058	0.2595	0.8292	-
6.0804	106000	0.0554	0.2790	0.8257	-
6.1091	106500	0.0593	0.2670	0.8270	-
6.1378	107000	0.0565	0.2732	0.8287	-
6.1665	107500	0.0596	0.2626	0.8271	-
6.1951	108000	0.0593	0.2639	0.8289	-
6.2238	108500	0.0655	0.2668	0.8234	-
6.2525	109000	0.0601	0.2649	0.8311	-
6.2812	109500	0.0653	0.2657	0.8270	-
6.3099	110000	0.06	0.2645	0.8246	-
6.3386	110500	0.0639	0.2665	0.8263	-
6.3672	111000	0.0553	0.2663	0.8259	-
6.3959	111500	0.0576	0.2859	0.8228	-
6.4246	112000	0.0567	0.2760	0.8270	-
6.4533	112500	0.059	0.2755	0.8235	-
6.4820	113000	0.0618	0.2662	0.8237	-
6.5106	113500	0.0569	0.2809	0.8270	-
6.5393	114000	0.0637	0.2725	0.8297	-
6.5680	114500	0.0584	0.2680	0.8269	-
6.5967	115000	0.0637	0.2639	0.8288	-
6.6254	115500	0.061	0.2734	0.8264	-
6.6540	116000	0.0632	0.2686	0.8273	-
6.6827	116500	0.0658	0.2743	0.8249	-
6.7114	117000	0.0594	0.2787	0.8310	-
6.7401	117500	0.0606	0.2760	0.8330	-
6.7688	118000	0.0589	0.2805	0.8256	-
6.7975	118500	0.057	0.2765	0.8349	-
6.8261	119000	0.0592	0.2648	0.8322	-
6.8548	119500	0.0568	0.2742	0.8323	-
6.8835	120000	0.0581	0.2679	0.8368	-
6.9122	120500	0.0565	0.2652	0.8328	-
6.9409	121000	0.0596	0.2756	0.8341	-
6.9695	121500	0.0622	0.2729	0.8303	-
6.9982	122000	0.0566	0.2742	0.8362	-
7.0269	122500	0.0516	0.2795	0.8302	-
7.0556	123000	0.0563	0.2730	0.8286	-
7.0843	123500	0.0489	0.2761	0.8326	-
7.1129	124000	0.0522	0.2721	0.8381	-
7.1416	124500	0.0505	0.2708	0.8313	-
7.1703	125000	0.054	0.2731	0.8332	-
7.1990	125500	0.0527	0.2814	0.8317	-
7.2277	126000	0.0557	0.2750	0.8299	-
7.2564	126500	0.0495	0.2781	0.8283	-
7.2850	127000	0.0491	0.2793	0.8263	-
7.3137	127500	0.0472	0.2842	0.8227	-
7.3424	128000	0.0482	0.2772	0.8270	-
7.3711	128500	0.0541	0.2723	0.8322	-
7.3998	129000	0.0512	0.2691	0.8348	-
7.4284	129500	0.0541	0.2704	0.8301	-
7.4571	130000	0.051	0.2677	0.8268	-
7.4858	130500	0.0517	0.2764	0.8248	-
7.5145	131000	0.0503	0.2839	0.8293	-
7.5432	131500	0.0495	0.2767	0.8247	-
7.5718	132000	0.0486	0.2777	0.8235	-
7.6005	132500	0.0497	0.2689	0.8289	-
7.6292	133000	0.0515	0.2773	0.8231	-
7.6579	133500	0.0489	0.2771	0.8262	-
7.6866	134000	0.0579	0.2713	0.8300	-
7.7153	134500	0.0508	0.2768	0.8298	-
7.7439	135000	0.0506	0.2726	0.8321	-
7.7726	135500	0.0516	0.2781	0.8274	-
7.8013	136000	0.0493	0.2805	0.8257	-
7.8300	136500	0.0453	0.2761	0.8269	-
7.8587	137000	0.0474	0.2774	0.8241	-
7.8873	137500	0.0482	0.2777	0.8233	-
7.9160	138000	0.0523	0.2803	0.8262	-
7.9447	138500	0.049	0.2714	0.8202	-
7.9734	139000	0.0504	0.2783	0.8263	-
8.0021	139500	0.0496	0.2751	0.8218	-
8.0307	140000	0.0412	0.2758	0.8229	-
8.0594	140500	0.0429	0.2825	0.8227	-
8.0881	141000	0.0427	0.2839	0.8243	-
8.1168	141500	0.0449	0.2816	0.8243	-
8.1455	142000	0.046	0.2733	0.8276	-
8.1742	142500	0.0382	0.2879	0.8229	-
8.2028	143000	0.0405	0.2794	0.8248	-
8.2315	143500	0.0468	0.2834	0.8244	-
8.2602	144000	0.0491	0.2751	0.8272	-
8.2889	144500	0.0498	0.2760	0.8308	-
8.3176	145000	0.0427	0.2818	0.8304	-
8.3462	145500	0.0432	0.2776	0.8297	-
8.3749	146000	0.0426	0.2766	0.8222	-
8.4036	146500	0.045	0.2793	0.8252	-
8.4323	147000	0.0417	0.2724	0.8274	-
8.4610	147500	0.0453	0.2721	0.8234	-
8.4896	148000	0.0472	0.2745	0.8249	-
8.5183	148500	0.0483	0.2719	0.8261	-
8.5470	149000	0.0452	0.2733	0.8286	-
8.5757	149500	0.0477	0.2742	0.8243	-
8.6044	150000	0.0414	0.2780	0.8260	-
8.6331	150500	0.0463	0.2740	0.8246	-
8.6617	151000	0.0476	0.2754	0.8250	-
8.6904	151500	0.045	0.2758	0.8249	-
8.7191	152000	0.0436	0.2844	0.8263	-
8.7478	152500	0.0413	0.2823	0.8231	-
8.7765	153000	0.0403	0.2812	0.8230	-
8.8051	153500	0.0444	0.2767	0.8251	-
8.8338	154000	0.0439	0.2773	0.8197	-
8.8625	154500	0.0431	0.2760	0.8199	-
8.8912	155000	0.0463	0.2676	0.8215	-
8.9199	155500	0.0409	0.2705	0.8216	-
8.9485	156000	0.0415	0.2711	0.8284	-
8.9772	156500	0.0437	0.2710	0.8233	-
9.0059	157000	0.0423	0.2774	0.8268	-
9.0346	157500	0.0357	0.2848	0.8247	-
9.0633	158000	0.0373	0.2838	0.8257	-
9.0920	158500	0.0356	0.2803	0.8259	-
9.1206	159000	0.0339	0.2815	0.8256	-
9.1493	159500	0.0405	0.2744	0.8244	-
9.1780	160000	0.0383	0.2769	0.8250	-
9.2067	160500	0.0397	0.2743	0.8259	-
9.2354	161000	0.0395	0.2843	0.8229	-
9.2640	161500	0.0354	0.2768	0.8266	-
9.2927	162000	0.0405	0.2811	0.8233	-
9.3214	162500	0.0361	0.2741	0.8248	-
9.3501	163000	0.0366	0.2761	0.8254	-
9.3788	163500	0.0368	0.2821	0.8255	-
9.4074	164000	0.0354	0.2790	0.8272	-
9.4361	164500	0.0365	0.2796	0.8239	-
9.4648	165000	0.0353	0.2763	0.8282	-
9.4935	165500	0.0378	0.2912	0.8206	-
9.5222	166000	0.0392	0.2801	0.8236	-
9.5509	166500	0.0402	0.2780	0.8275	-
9.5795	167000	0.0395	0.2805	0.8217	-
9.6082	167500	0.0366	0.2768	0.8218	-
9.6369	168000	0.0364	0.2719	0.8242	-
9.6656	168500	0.0361	0.2748	0.8220	-
9.6943	169000	0.0345	0.2757	0.8212	-
9.7229	169500	0.0396	0.2801	0.8242	-
9.7516	170000	0.0359	0.2814	0.8246	-
9.7803	170500	0.039	0.2765	0.8275	-
9.8090	171000	0.0372	0.2786	0.8245	-
9.8377	171500	0.0347	0.2735	0.8273	-
9.8663	172000	0.0376	0.2800	0.8234	-
9.8950	172500	0.0336	0.2784	0.8255	-
9.9237	173000	0.0326	0.2754	0.8241	-
9.9524	173500	0.0373	0.2768	0.8217	-
9.9811	174000	0.0393	0.2757	0.8207	-
10.0098	174500	0.0341	0.2720	0.8201	-
10.0384	175000	0.0316	0.2762	0.8220	-
10.0671	175500	0.032	0.2791	0.8234	-
10.0958	176000	0.0343	0.2807	0.8204	-
10.1245	176500	0.0375	0.2780	0.8182	-
10.1532	177000	0.0368	0.2867	0.8223	-
10.1818	177500	0.0335	0.2812	0.8221	-
10.2105	178000	0.0345	0.2817	0.8214	-
10.2392	178500	0.0338	0.2851	0.8216	-
10.2679	179000	0.0365	0.2808	0.8284	-
10.2966	179500	0.0311	0.2774	0.8251	-
10.3252	180000	0.037	0.2782	0.8231	-
10.3539	180500	0.0347	0.2769	0.8236	-
10.3826	181000	0.0389	0.2841	0.8219	-
10.4113	181500	0.0345	0.2798	0.8224	-
10.4400	182000	0.0364	0.2890	0.8236	-
10.4687	182500	0.0332	0.2862	0.8217	-
10.4973	183000	0.0305	0.2828	0.8196	-
10.5260	183500	0.0312	0.2781	0.8237	-
10.5547	184000	0.0305	0.2794	0.8250	-
10.5834	184500	0.0336	0.2858	0.8226	-
10.6121	185000	0.0324	0.2834	0.8290	-
10.6407	185500	0.0343	0.2869	0.8282	-
10.6694	186000	0.0345	0.2825	0.8280	-
10.6981	186500	0.033	0.2960	0.8245	-
10.7268	187000	0.0327	0.2791	0.8279	-
10.7555	187500	0.0318	0.2821	0.8253	-
10.7841	188000	0.0346	0.2874	0.8241	-
10.8128	188500	0.0309	0.2857	0.8229	-
10.8415	189000	0.0355	0.2815	0.8234	-
10.8702	189500	0.0318	0.2839	0.8235	-
10.8989	190000	0.0307	0.2864	0.8242	-
10.9276	190500	0.0325	0.2792	0.8250	-
10.9562	191000	0.0314	0.2773	0.8269	-
10.9849	191500	0.0377	0.2877	0.8257	-
11.0136	192000	0.0267	0.2892	0.8213	-
11.0423	192500	0.028	0.2905	0.8196	-
11.0710	193000	0.0309	0.2894	0.8224	-
11.0996	193500	0.031	0.2860	0.8219	-
11.1283	194000	0.0333	0.2868	0.8190	-
11.1570	194500	0.0305	0.2886	0.8205	-
11.1857	195000	0.0345	0.2903	0.8214	-
11.2144	195500	0.0324	0.2936	0.8228	-
11.2430	196000	0.0315	0.2963	0.8218	-
11.2717	196500	0.0295	0.2997	0.8205	-
11.3004	197000	0.0297	0.2939	0.8236	-
11.3291	197500	0.0293	0.2929	0.8214	-
11.3578	198000	0.0345	0.2937	0.8198	-
11.3865	198500	0.0317	0.2942	0.8220	-
11.4151	199000	0.0329	0.2963	0.8244	-
11.4438	199500	0.033	0.2909	0.8254	-
11.4725	200000	0.0293	0.2983	0.8234	-
11.5012	200500	0.0315	0.3012	0.8229	-
11.5299	201000	0.0294	0.2972	0.8225	-
11.5585	201500	0.0329	0.2940	0.8228	-
11.5872	202000	0.0331	0.2858	0.8226	-
11.6159	202500	0.028	0.2891	0.8240	-
11.6446	203000	0.0273	0.2936	0.8212	-
11.6733	203500	0.0288	0.2935	0.8193	-
11.7019	204000	0.0291	0.2892	0.8195	-
11.7306	204500	0.0315	0.2828	0.8195	-
11.7593	205000	0.0296	0.2928	0.8192	-
11.7880	205500	0.0251	0.2909	0.8211	-
11.8167	206000	0.029	0.2938	0.8218	-
11.8454	206500	0.0293	0.2877	0.8224	-
11.8740	207000	0.0295	0.2835	0.8227	-
11.9027	207500	0.0252	0.2866	0.8227	-
11.9314	208000	0.0305	0.2866	0.8224	-
11.9601	208500	0.029	0.2909	0.8211	-
11.9888	209000	0.0258	0.2903	0.8216	-
12.0174	209500	0.0282	0.2865	0.8224	-
12.0461	210000	0.0248	0.2928	0.8193	-
12.0748	210500	0.0295	0.2895	0.8206	-
12.1035	211000	0.0245	0.2883	0.8200	-
12.1322	211500	0.0261	0.2888	0.8209	-
12.1608	212000	0.0262	0.2898	0.8176	-
12.1895	212500	0.0304	0.2887	0.8193	-
12.2182	213000	0.0245	0.2867	0.8179	-
12.2469	213500	0.0239	0.2872	0.8170	-
12.2756	214000	0.0257	0.2893	0.8183	-
12.3043	214500	0.0271	0.2889	0.8166	-
12.3329	215000	0.0268	0.2865	0.8161	-
12.3616	215500	0.0296	0.2908	0.8167	-
12.3903	216000	0.0254	0.2921	0.8171	-
12.4190	216500	0.0273	0.2890	0.8179	-
12.4477	217000	0.0269	0.2919	0.8187	-
12.4763	217500	0.0272	0.2919	0.8197	-
12.5050	218000	0.0289	0.2922	0.8207	-
12.5337	218500	0.0266	0.2935	0.8215	-
12.5624	219000	0.0262	0.2977	0.8192	-
12.5911	219500	0.0235	0.2920	0.8204	-
12.6197	220000	0.03	0.2899	0.8206	-
12.6484	220500	0.0251	0.2935	0.8175	-
12.6771	221000	0.0295	0.2923	0.8197	-
12.7058	221500	0.0264	0.2955	0.8180	-
12.7345	222000	0.0273	0.2952	0.8186	-
12.7632	222500	0.0261	0.2927	0.8203	-
12.7918	223000	0.0282	0.2924	0.8194	-
12.8205	223500	0.0263	0.2946	0.8215	-
12.8492	224000	0.0263	0.2949	0.8205	-
12.8779	224500	0.0255	0.2923	0.8214	-
12.9066	225000	0.0263	0.2900	0.8202	-
12.9352	225500	0.0266	0.2956	0.8195	-
12.9639	226000	0.0229	0.2897	0.8211	-
12.9926	226500	0.0291	0.2901	0.8214	-
13.0213	227000	0.0256	0.2921	0.8212	-
13.0500	227500	0.0232	0.2910	0.8211	-
13.0786	228000	0.0237	0.2961	0.8209	-
13.1073	228500	0.0257	0.2987	0.8182	-
13.1360	229000	0.0215	0.3004	0.8167	-
13.1647	229500	0.0269	0.2984	0.8168	-
13.1934	230000	0.0263	0.2963	0.8182	-
13.2221	230500	0.0263	0.2967	0.8172	-
13.2507	231000	0.0257	0.2927	0.8186	-
13.2794	231500	0.0234	0.2976	0.8207	-
13.3081	232000	0.0244	0.2980	0.8200	-
13.3368	232500	0.0272	0.2934	0.8199	-
13.3655	233000	0.0228	0.2942	0.8203	-
13.3941	233500	0.0259	0.2988	0.8209	-
13.4228	234000	0.0223	0.2993	0.8191	-
13.4515	234500	0.0235	0.2947	0.8203	-
13.4802	235000	0.0221	0.2968	0.8200	-
13.5089	235500	0.023	0.3007	0.8197	-
13.5375	236000	0.0246	0.2941	0.8200	-
13.5662	236500	0.0254	0.2966	0.8187	-
13.5949	237000	0.0244	0.2969	0.8175	-
13.6236	237500	0.0257	0.2998	0.8184	-
13.6523	238000	0.0261	0.3003	0.8173	-
13.6809	238500	0.0232	0.2991	0.8176	-
13.7096	239000	0.0248	0.2998	0.8182	-
13.7383	239500	0.0226	0.3012	0.8170	-
13.7670	240000	0.0252	0.2979	0.8179	-
13.7957	240500	0.0254	0.2957	0.8197	-
13.8244	241000	0.0228	0.2976	0.8203	-
13.8530	241500	0.0226	0.2941	0.8198	-
13.8817	242000	0.0225	0.2970	0.8189	-
13.9104	242500	0.023	0.3030	0.8182	-
13.9391	243000	0.0237	0.3022	0.8188	-
13.9678	243500	0.0231	0.2984	0.8197	-
13.9964	244000	0.0239	0.2976	0.8188	-
14.0251	244500	0.0232	0.2985	0.8191	-
14.0538	245000	0.0237	0.2969	0.8196	-
14.0825	245500	0.0214	0.2964	0.8205	-
14.1112	246000	0.024	0.2971	0.8198	-
14.1398	246500	0.0221	0.2982	0.8189	-
14.1685	247000	0.0186	0.2994	0.8188	-
14.1972	247500	0.0242	0.2965	0.8193	-
14.2259	248000	0.0238	0.2992	0.8193	-
14.2546	248500	0.0228	0.2989	0.8188	-
14.2833	249000	0.0235	0.2983	0.8188	-
14.3119	249500	0.026	0.2999	0.8187	-
14.3406	250000	0.0189	0.2996	0.8185	-
14.3693	250500	0.0211	0.3023	0.8183	-
14.3980	251000	0.0232	0.3030	0.8176	-
14.4267	251500	0.0216	0.3018	0.8177	-
14.4553	252000	0.0236	0.3012	0.8184	-
14.4840	252500	0.0201	0.3017	0.8187	-
14.5127	253000	0.019	0.3020	0.8184	-
14.5414	253500	0.0208	0.3027	0.8180	-
14.5701	254000	0.0223	0.3016	0.8179	-
14.5987	254500	0.0205	0.3028	0.8176	-
14.6274	255000	0.0227	0.3018	0.8180	-
14.6561	255500	0.022	0.3031	0.8175	-
14.6848	256000	0.0228	0.3016	0.8173	-
14.7135	256500	0.024	0.3010	0.8175	-
14.7422	257000	0.0228	0.3015	0.8176	-
14.7708	257500	0.0235	0.3012	0.8178	-
14.7995	258000	0.0225	0.3008	0.8180	-
14.8282	258500	0.0173	0.3013	0.8178	-
14.8569	259000	0.0215	0.3010	0.8180	-
14.8856	259500	0.0219	0.3011	0.8179	-
14.9142	260000	0.022	0.3011	0.8178	-
14.9429	260500	0.023	0.3012	0.8177	-
14.9716	261000	0.0234	0.3012	0.8177	-
-1	-1	-	-	-	0.7887

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}
}