dnth/ssf-retriever-modernbert-embed-base

SentenceTransformer based on Qwen/Qwen3-Embedding-0.6B

This is a sentence-transformers model finetuned from Qwen/Qwen3-Embedding-0.6B. It maps sentences & paragraphs to a 1024-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: Qwen/Qwen3-Embedding-0.6B
• Maximum Sequence Length: 32768 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 32768, 'do_lower_case': False, 'architecture': 'Qwen3Model'})
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, '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': True, '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("dnth/ssf-retriever-modernbert-embed-base")
# Run inference
queries = [
    "The Principal Psychologist Educator develops and delivers educational programmes in psychology and works in collaboration with professionals from direct practice and external organisations across sectors to develop training curricula, programmes and delivery methods for effective training delivery. He/She also facilitates the improvement and development of new educational services and supports capability development within the department and at an organisational level. He supervises and mentors junior staff in the delivery of educational programmes in psychology. He also works with professionals from direct practice and research to conceptualise and conduct education-related research. An experienced professional who possesses strong facilitation and communication skills, the Principal Psychologist Educator is collaborative in his approach and works in varied settings such as ministries, public and private institutions, hospitals, healthcare and voluntary welfare organisations.",
]
documents = [
    'Educational Programme Developer in Psychology responsible for designing and implementing psychology training initiatives, collaborating with various professionals to create effective curricula and delivery methods, while enhancing educational services and supporting skill development across the organization. This role involves mentoring junior educators and conducting research related to educational practices in psychology within diverse settings including healthcare, public sectors, and educational institutions.',
    'Junior Financial Analyst focused on preparing and analyzing financial reports, working closely with various departments to ensure accurate data collection and reporting methods. This role involves supporting the development of financial strategies and assisting in the implementation of budgeting processes while collaborating with team members to improve overall financial performance. The Junior Financial Analyst also engages in research related to financial trends and market analysis within corporate sectors and non-profit organizations.',
    'Food Scientist specializing in the development of innovative and nutritious food products, leveraging food science principles to explore alternative ingredients and processing techniques while focusing on market trends and consumer needs. Responsible for managing labs and pilot plants to enhance production scalability and ensure compliance with safety standards.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 1024] [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.7510, 0.0014, 0.0243]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 7,540 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 53 tokens mean: 163.18 tokens max: 394 tokens	min: 19 tokens mean: 62.74 tokens max: 211 tokens	min: 33 tokens mean: 78.3 tokens max: 230 tokens

• Samples:

  anchor	positive	negative
  The Head Baker leads the preparation of a variety of baked goods. He/She inspects the ingredients used for daily products and the finishing touches of baked goods. He also performs audits on staffs compliance with hygiene, safety and other standards, and suggests areas for continuous improvement within the team. He is expected to provide recommendations in the development of new recipes to renew menus. Meticulous and resourceful, he possesses mental resilience to operate in high pressure environments, and is capable in communicating and working effectively with co-workers and suppliers. He should be comfortable with standing for long hours to monitor the baking process. He is expected to manage competing priorities and multiple deadlines in a fast-paced environment.	Lead Baker overseeing the creation of diverse baked items, ensuring ingredient quality and presentation. Conducts staff hygiene and safety compliance audits while identifying areas for team improvement. Responsible for proposing new recipes to enhance the menu, demonstrating attention to detail and resourcefulness. Must thrive in high-pressure settings and effectively collaborate with team members and suppliers, with the ability to manage multiple tasks while standing for extended periods.	Junior Pastry Chef responsible for assisting in the preparation of desserts and pastries in a busy restaurant. This role involves managing inventory levels of ingredients and ensuring compliance with kitchen safety standards. The candidate will work under the supervision of the Executive Chef, focusing on executing daily dessert specials and maintaining cleanliness in the kitchen. Must be able to handle feedback and work collaboratively with the kitchen staff, while adapting to a dynamic culinary environment.
  The Business Development Director assumes overall responsibility for leading all business development efforts within the organisation, including the development and implementation of business development strategies and activities. Through expansion of current businesses and exploration of new markets and opportunities, he/she spearheads business growth for the organisation. He also leads business development activities through cross-function collaborations. Through partnerships, Joint Ventures (JV) and Mergers and Acquisitions (M&A), he endeavours to grow and expand the market share of the organisation. Assertive and insightful, he possesses strong business acumen and entrepreneurial instinct that enables him to source for growth opportunities. He keeps abreast of market trends, industry events, competitors actions and clients' needs in order to be pro-active in pursuing growth opportunities. He is able to respond quickly to improve the effectiveness of current plans and programmes to ...	Business Development Manager responsible for driving strategic initiatives and spearheading growth through market expansion and partnership development, while collaborating across functions to enhance business opportunities and client relationships.	Junior Marketing Coordinator needed to assist in executing promotional campaigns and managing social media content for a retail company. The role involves coordinating with various teams to create engaging marketing materials, analyzing customer feedback, and supporting the marketing manager in daily operations. Strong communication skills and creativity are essential, along with a good understanding of market trends and consumer behavior.
  The Senior Server Programmer leads the design and development of online game server networks to support various game features such as online gameplay, in-game events and purchases, credential verification and online messaging systems. He/She is responsible for translating the vision for online features into a server network design and realising it by configuring appropriate hardware. He oversees the development of programs to enable the game to interact with the servers. He reviews server programs, oversees the testing of online gameplay features and leads the integration of server programs within the overall game code. He also oversees the maintenance of game servers and online operations. The role involves leading a team of programmers with technical guidance as well as liaising with other teams, internal and external stakeholders to ensure project expectations are met. He also spends a significant amount of his time in meetings with other production teams to align expectations and s...	Lead the design and development of online game server networks, focusing on gameplay features, in-game events, and online interactions while managing a team of programmers and collaborating with cross-functional teams.	The Junior Business Analyst is responsible for supporting the assessment and analysis of business processes and systems within the healthcare industry. This role involves gathering and documenting business requirements, assisting in the development of project plans, and collaborating with stakeholders to ensure project deliverables align with organizational goals. The Junior Business Analyst will also conduct research on industry trends and assist in the implementation of new systems and processes. Strong analytical skills and proficiency in data analysis tools are essential for this role, along with effective communication skills to liaise with various departments.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 1,885 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 56 tokens mean: 161.67 tokens max: 364 tokens	min: 18 tokens mean: 62.19 tokens max: 171 tokens	min: 30 tokens mean: 77.63 tokens max: 182 tokens

• Samples:

  anchor	positive	negative
  The Restructuring and Insolvency Senior/Restructuring and Insolvency Senior Executive is in charge of day-to-day operations, from a restructuring and insolvency perspective. He/She manages the restructuring and insolvency processes for the client engagements that he is responsible for, or the business that he belongs to. He is expected to adhere to standards of ethics and maintain quality assurance in processes. He participates in business development and is expected to interact with stakeholders to manage project deliverables and timelines. He has a significant level of technical expertise and is very hands-on with the restructuring and insolvency processes. He must be able to work in a fast-paced environment. He needs to have strong project management skills and be efficient in his work to manage multiple deadlines. He is able to interpret data and communicate the insights derived to his team members.	Restructuring and Insolvency Executive responsible for managing client engagements, overseeing daily operations, ensuring adherence to ethical standards, and maintaining quality assurance while interacting with stakeholders and managing project timelines.	Junior Financial Analyst in the healthcare industry tasked with supporting senior analysts in evaluating financial data, preparing reports on budget forecasts, and assisting with compliance audits. This role requires proficiency in Excel and strong analytical skills, focusing on data interpretation and communication within a team setting.
  The Senior Technician (Mechanical and Electrical) performs preventive and corrective maintenance of mechanical and electrical systems. He/She is technically inclined, knowledgeable and skilled in the maintenance of various mechanical and electrical systems His duties include troubleshooting faults, providing technical guidance and on-the-job coaching to his team, as well as supervising the work of contractors and external stakeholders in ensuring compliance to safety requirements and operating standards. He is required to work in shifts and carries out his duties at various rail premises such as workshops and at train stations. He is a team-player and is able to communicate effectively within the team to support maintenance activities.	Seeking a Mechanical and Electrical Maintenance Technician to conduct preventive and corrective maintenance on various systems. The ideal candidate will possess strong technical skills, be knowledgeable in troubleshooting and providing guidance, and supervise contractors to ensure safety compliance. Shift work is required at rail facilities, including workshops and train stations, with a focus on teamwork and effective communication.	Looking for a Junior Financial Analyst to assist with risk assessments and financial reporting in the healthcare sector. The candidate should be familiar with financial modeling and data analysis, providing support to senior analysts and collaborating with cross-functional teams. Responsibilities include preparing reports, analyzing financial data, and ensuring compliance with industry regulations. Strong communication skills are essential for this role.
  The Customer Success Director is responsible for establishing strategies to drive customer satisfaction to increase retention and lifetime value for the organisation. He/She defines critical success factors for the team and provides advice on the development of client onboarding, engagement initiatives and programs to ensure successful adoption of solutions and realisation of optimal value for the client. He oversees the development of educational resources and case studies, as well as recommendations and action plans to address challenges faced by the client. He leverages relationships with clients to drive opportunities for new business developments and up-selling and cross-selling. He works in a fast-paced and dynamic environment, and visits clients' premises as and when required. He is familiar with client relationship management and sales tools, as well as customer service frameworks and practices. He is knowledgeable of best practices pertaining to the use of the organisation's p...	Customer Success Manager focused on enhancing client satisfaction and loyalty by implementing strategies for retention and maximizing lifetime value. Responsible for defining success metrics, guiding onboarding processes, and developing engagement programs to ensure clients derive maximum value from our solutions. Oversees the creation of educational materials and case studies while providing actionable insights to tackle client challenges. Utilizes client relationships to explore new business opportunities and facilitate upselling and cross-selling. Works in a dynamic environment with occasional client visits. Proficient in CRM and sales tools, as well as customer service standards. Well-versed in best practices for product usage and knowledgeable about industry-specific business needs. The role requires strong analytical skills and a proactive approach to market trends and changes, along with excellent leadership and interpersonal skills to influence stakeholders and mentor team memb...	Junior Customer Service Representative tasked with handling customer inquiries and resolving issues to maintain satisfaction levels. Responsible for processing orders, managing returns, and providing product information to clients. Engages in routine follow-ups with customers to ensure service quality and address any concerns that may arise. Works under supervision to understand customer needs and escalate complex issues to senior staff. Familiarity with basic customer service protocols and communication tools is essential. The role focuses on maintaining a positive customer experience through efficient problem-solving and support. Requires strong communication skills and the ability to work in a team-oriented environment, with a focus on meeting service level agreements and targets.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 4
• per_device_eval_batch_size: 2
• gradient_accumulation_steps: 2
• learning_rate: 2e-05
• num_train_epochs: 4
• lr_scheduler_type: cosine
• warmup_ratio: 0.1
• bf16: True
• tf32: True
• load_best_model_at_end: True
• optim: adamw_torch_fused
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 4
• per_device_eval_batch_size: 2
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 2
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 2e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 4
• max_steps: -1
• lr_scheduler_type: cosine
• 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: True
• fp16: False
• fp16_opt_level: O1
• half_precision_backend: auto
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: True
• 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_fused
• 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
• 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: False
• 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: False
• prompts: None
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss
0.0106	10	0.0097	-
0.0212	20	0.0089	-
0.0318	30	0.0051	-
0.0424	40	0.0015	-
0.0531	50	0.0005	-
0.0637	60	0.0005	-
0.0743	70	0.0001	-
0.0849	80	0.0002	-
0.0955	90	0.0002	-
0.1061	100	0.0022	-
0.1167	110	0.0013	-
0.1273	120	0.0026	-
0.1379	130	0.0007	-
0.1485	140	0.0002	-
0.1592	150	0.0001	-
0.1698	160	0.0013	-
0.1804	170	0.0002	-
0.1910	180	0.0001	-
0.2016	190	0.0003	-
0.2122	200	0.0063	-
0.2228	210	0.0003	-
0.2334	220	0.0002	-
0.2440	230	0.0074	-
0.2546	240	0.0002	-
0.2653	250	0.0001	-
0.2759	260	0.0001	-
0.2865	270	0.001	-
0.2971	280	0.0337	-
0.3077	290	0.0004	-
0.3183	300	0.0001	-
0.3289	310	0.0033	-
0.3395	320	0.0003	-
0.3501	330	0.0094	-
0.3607	340	0.0027	-
0.3714	350	0.0052	-
0.3820	360	0.0011	-
0.3926	370	0.0007	-
0.4032	380	0.0001	-
0.4138	390	0.0005	-
0.4244	400	0.0001	-
0.4350	410	0.0001	-
0.4456	420	0.0001	-
0.4562	430	0.0001	-
0.4668	440	0.0003	-
0.4775	450	0.0	-
0.4881	460	0.0079	-
0.4987	470	0.0005	-
0.5093	480	0.0024	-
0.5199	490	0.0008	-
0.5305	500	0.0027	-
0.5411	510	0.0046	-
0.5517	520	0.0003	-
0.5623	530	0.0019	-
0.5729	540	0.0005	-
0.5836	550	0.0092	-
0.5942	560	0.0006	-
0.6048	570	0.0014	-
0.6154	580	0.0009	-
0.6260	590	0.0005	-
0.6366	600	0.0003	-
0.6472	610	0.0002	-
0.6578	620	0.0005	-
0.6684	630	0.0001	-
0.6790	640	0.0003	-
0.6897	650	0.0047	-
0.7003	660	0.0002	-
0.7109	670	0.0002	-
0.7215	680	0.0001	-
0.7321	690	0.0006	-
0.7427	700	0.0004	-
0.7533	710	0.0002	-
0.7639	720	0.0002	-
0.7745	730	0.0073	-
0.7851	740	0.0001	-
0.7958	750	0.0031	-
0.8064	760	0.0037	-
0.8170	770	0.0018	-
0.8276	780	0.0002	-
0.8382	790	0.0018	-
0.8488	800	0.0399	-
0.8594	810	0.0199	-
0.8700	820	0.0431	-
0.8806	830	0.032	-
0.8912	840	0.0019	-
0.9019	850	0.0029	-
0.9125	860	0.0255	-
0.9231	870	0.0112	-
0.9337	880	0.012	-
0.9443	890	0.0028	-
0.9549	900	0.0331	-
0.9655	910	0.0012	-
0.9761	920	0.0005	-
0.9867	930	0.0011	-
0.9973	940	0.0077	-
1.0	943	-	0.0012
1.0074	950	0.0018	-
1.0180	960	0.0028	-
1.0286	970	0.001	-
1.0393	980	0.0009	-
1.0499	990	0.0054	-
1.0605	1000	0.0004	-
1.0711	1010	0.0021	-
1.0817	1020	0.0012	-
1.0923	1030	0.0041	-
1.1029	1040	0.0018	-
1.1135	1050	0.0008	-
1.1241	1060	0.0007	-
1.1347	1070	0.004	-
1.1454	1080	0.0003	-
1.1560	1090	0.0002	-
1.1666	1100	0.0001	-
1.1772	1110	0.0001	-
1.1878	1120	0.0067	-
1.1984	1130	0.0003	-
1.2090	1140	0.0015	-
1.2196	1150	0.0004	-
1.2302	1160	0.0008	-
1.2408	1170	0.0004	-
1.2515	1180	0.0001	-
1.2621	1190	0.0015	-
1.2727	1200	0.0017	-
1.2833	1210	0.0001	-
1.2939	1220	0.019	-
1.3045	1230	0.0036	-
1.3151	1240	0.0003	-
1.3257	1250	0.0395	-
1.3363	1260	0.0226	-
1.3469	1270	0.0005	-
1.3576	1280	0.0056	-
1.3682	1290	0.0002	-
1.3788	1300	0.0008	-
1.3894	1310	0.0011	-
1.4	1320	0.0144	-
1.4106	1330	0.0012	-
1.4212	1340	0.0005	-
1.4318	1350	0.0016	-
1.4424	1360	0.0051	-
1.4531	1370	0.0022	-
1.4637	1380	0.0061	-
1.4743	1390	0.003	-
1.4849	1400	0.0011	-
1.4955	1410	0.0298	-
1.5061	1420	0.0004	-
1.5167	1430	0.0001	-
1.5273	1440	0.0001	-
1.5379	1450	0.0041	-
1.5485	1460	0.0045	-
1.5592	1470	0.0001	-
1.5698	1480	0.0003	-
1.5804	1490	0.0002	-
1.5910	1500	0.0124	-
1.6016	1510	0.0005	-
1.6122	1520	0.0003	-
1.6228	1530	0.0005	-
1.6334	1540	0.0006	-
1.6440	1550	0.0004	-
1.6546	1560	0.0002	-
1.6653	1570	0.0005	-
1.6759	1580	0.0012	-
1.6865	1590	0.0001	-
1.6971	1600	0.0002	-
1.7077	1610	0.0118	-
1.7183	1620	0.0005	-
1.7289	1630	0.0009	-
1.7395	1640	0.0026	-
1.7501	1650	0.0079	-
1.7607	1660	0.0011	-
1.7714	1670	0.0002	-
1.7820	1680	0.0006	-
1.7926	1690	0.0001	-
1.8032	1700	0.0006	-
1.8138	1710	0.0004	-
1.8244	1720	0.0001	-
1.8350	1730	0.0012	-
1.8456	1740	0.0015	-
1.8562	1750	0.0002	-
1.8668	1760	0.0004	-
1.8775	1770	0.0013	-
1.8881	1780	0.0	-
1.8987	1790	0.001	-
1.9093	1800	0.0003	-
1.9199	1810	0.0007	-
1.9305	1820	0.0005	-
1.9411	1830	0.001	-
1.9517	1840	0.0059	-
1.9623	1850	0.0001	-
1.9729	1860	0.0003	-
1.9836	1870	0.0002	-
1.9942	1880	0.0022	-
2.0	1886	-	0.0002
2.0042	1890	0.0003	-
2.0149	1900	0.0	-
2.0255	1910	0.0089	-
2.0361	1920	0.0001	-
2.0467	1930	0.0001	-
2.0573	1940	0.0002	-
2.0679	1950	0.0006	-
2.0785	1960	0.0004	-
2.0891	1970	0.0001	-
2.0997	1980	0.0001	-
2.1103	1990	0.0002	-
2.1210	2000	0.0001	-
2.1316	2010	0.0064	-
2.1422	2020	0.0004	-
2.1528	2030	0.0003	-
2.1634	2040	0.0002	-
2.1740	2050	0.0004	-
2.1846	2060	0.0	-
2.1952	2070	0.0001	-
2.2058	2080	0.0001	-
2.2164	2090	0.0002	-
2.2271	2100	0.0004	-
2.2377	2110	0.0003	-
2.2483	2120	0.0001	-
2.2589	2130	0.0001	-
2.2695	2140	0.0004	-
2.2801	2150	0.0002	-
2.2907	2160	0.0007	-
2.3013	2170	0.0004	-
2.3119	2180	0.0003	-
2.3225	2190	0.0001	-
2.3332	2200	0.0001	-
2.3438	2210	0.0	-
2.3544	2220	0.0001	-
2.3650	2230	0.0001	-
2.3756	2240	0.0005	-
2.3862	2250	0.0001	-
2.3968	2260	0.0001	-
2.4074	2270	0.0	-
2.4180	2280	0.0003	-
2.4286	2290	0.0023	-
2.4393	2300	0.001	-
2.4499	2310	0.0006	-
2.4605	2320	0.0004	-
2.4711	2330	0.0001	-
2.4817	2340	0.0281	-
2.4923	2350	0.0001	-
2.5029	2360	0.0016	-
2.5135	2370	0.0002	-
2.5241	2380	0.0019	-
2.5347	2390	0.0001	-
2.5454	2400	0.0001	-
2.5560	2410	0.0002	-
2.5666	2420	0.0	-
2.5772	2430	0.0001	-
2.5878	2440	0.0001	-
2.5984	2450	0.0029	-
2.6090	2460	0.0094	-
2.6196	2470	0.0001	-
2.6302	2480	0.0005	-
2.6408	2490	0.0001	-
2.6515	2500	0.0	-
2.6621	2510	0.0003	-
2.6727	2520	0.0001	-
2.6833	2530	0.0001	-
2.6939	2540	0.0012	-
2.7045	2550	0.0001	-
2.7151	2560	0.0001	-
2.7257	2570	0.0002	-
2.7363	2580	0.0	-
2.7469	2590	0.0003	-
2.7576	2600	0.0001	-
2.7682	2610	0.0	-
2.7788	2620	0.0001	-
2.7894	2630	0.0003	-
2.8	2640	0.0032	-
2.8106	2650	0.0001	-
2.8212	2660	0.0001	-
2.8318	2670	0.0	-
2.8424	2680	0.0023	-
2.8531	2690	0.0001	-
2.8637	2700	0.0001	-
2.8743	2710	0.0009	-
2.8849	2720	0.0006	-
2.8955	2730	0.0001	-
2.9061	2740	0.0	-
2.9167	2750	0.001	-
2.9273	2760	0.0	-
2.9379	2770	0.0003	-
2.9485	2780	0.0	-
2.9592	2790	0.0	-
2.9698	2800	0.0	-
2.9804	2810	0.0001	-
2.9910	2820	0.0001	-
3.0	2829	-	0.0001
3.0011	2830	0.0	-
3.0117	2840	0.0001	-
3.0223	2850	0.0001	-
3.0329	2860	0.0	-
3.0435	2870	0.0	-
3.0541	2880	0.0001	-
3.0647	2890	0.0078	-
3.0753	2900	0.0012	-
3.0859	2910	0.0	-
3.0966	2920	0.0006	-
3.1072	2930	0.0001	-
3.1178	2940	0.0	-
3.1284	2950	0.0001	-
3.1390	2960	0.0	-
3.1496	2970	0.0	-
3.1602	2980	0.0001	-
3.1708	2990	0.0016	-
3.1814	3000	0.0	-
3.1920	3010	0.0001	-
3.2027	3020	0.0002	-
3.2133	3030	0.0	-
3.2239	3040	0.0	-
3.2345	3050	0.0013	-
3.2451	3060	0.0003	-
3.2557	3070	0.0003	-
3.2663	3080	0.0	-
3.2769	3090	0.0	-
3.2875	3100	0.0	-
3.2981	3110	0.0	-
3.3088	3120	0.0001	-
3.3194	3130	0.0003	-
3.3300	3140	0.0001	-
3.3406	3150	0.0	-
3.3512	3160	0.0001	-
3.3618	3170	0.0001	-
3.3724	3180	0.0001	-
3.3830	3190	0.0004	-
3.3936	3200	0.0103	-
3.4042	3210	0.0001	-
3.4149	3220	0.0	-
3.4255	3230	0.0001	-
3.4361	3240	0.0033	-
3.4467	3250	0.0001	-
3.4573	3260	0.0	-
3.4679	3270	0.0	-
3.4785	3280	0.0004	-
3.4891	3290	0.0	-
3.4997	3300	0.0001	-
3.5103	3310	0.0101	-
3.5210	3320	0.0	-
3.5316	3330	0.0	-
3.5422	3340	0.0	-
3.5528	3350	0.0001	-
3.5634	3360	0.0	-
3.5740	3370	0.0	-
3.5846	3380	0.0001	-
3.5952	3390	0.0003	-
3.6058	3400	0.0	-
3.6164	3410	0.0028	-
3.6271	3420	0.0002	-
3.6377	3430	0.0001	-
3.6483	3440	0.0002	-
3.6589	3450	0.0001	-
3.6695	3460	0.0	-
3.6801	3470	0.0	-
3.6907	3480	0.0004	-
3.7013	3490	0.0001	-
3.7119	3500	0.0001	-
3.7225	3510	0.0	-
3.7332	3520	0.0001	-
3.7438	3530	0.0	-
3.7544	3540	0.0003	-
3.7650	3550	0.0001	-
3.7756	3560	0.0001	-
3.7862	3570	0.0009	-
3.7968	3580	0.0024	-
3.8074	3590	0.0002	-
3.8180	3600	0.0001	-
3.8286	3610	0.0001	-
3.8393	3620	0.0001	-
3.8499	3630	0.0	-
3.8605	3640	0.0	-
3.8711	3650	0.0	-
3.8817	3660	0.0005	-
3.8923	3670	0.0006	-
3.9029	3680	0.0002	-
3.9135	3690	0.0001	-
3.9241	3700	0.0015	-
3.9347	3710	0.0004	-
3.9454	3720	0.0	-
3.9560	3730	0.0	-
3.9666	3740	0.0	-
3.9772	3750	0.0	-
3.9878	3760	0.0	-
3.9984	3770	0.0	-
4.0	3772	-	0.0001

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.12.8
• Sentence Transformers: 5.0.0
• Transformers: 4.54.1
• PyTorch: 2.7.1+cu126
• Accelerate: 1.9.0
• Datasets: 4.0.0
• Tokenizers: 0.21.4

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",
}

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}
}