projfinetuned

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2. It maps sentences & paragraphs to a 384-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: sentence-transformers/all-MiniLM-L6-v2 <!-- at revision c9745ed1d9f207416be6d2e6f8de32d1f16199bf -->
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity

<!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

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': 256, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 384, '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})
  (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("its-zion-18/projfinetuned")
# Run inference
sentences = [
    'Manager, Social Sales & Newsletters',
    'Manager, Social Sales & Newsletters  Collaborate with the sales team to drive revenue and sales for social advertising business, ensuring revenue goals are hit quarterly and annually.\nManage day-to-day sales requests providing targeting information and inventory requests to sales planners & reps to assist in their pre-sale initiatives\nCollaborate with the sales team to drive revenue and sales for newsletter partnerships, ensuring revenue goals are hit quarterly and annually.\nAttend live events and partner with stakeholders to ensure contracted social elements for clients are operating smoothly when necessary\nManage communications and organization from pre-sale to post-sale including managing tracker of programs, working on brainstorms in pre-sale, joining kick off calls in post-sale, looping in social activation and newsletter teams and providing campaign details\nStrong understanding of social media best practices, trends and advertising on social platforms\nWe are holding ourselves accountable and ensuring that all decisions are made with equity top of mind. Bachelor’s Degree in Business, Marketing, Communications or a related field\n3-5 years of relevant experience\nCreative storyteller\nCreator economy experience a must\nStrong proficiency in Excel, Google Drive\nAbility to multi-task, efficiently manage time and prioritize deliverables\nForbes has estimated the compensation range set forth above in good faith.\xa0 The compensation range is what we believe we will offer, and ultimately pay, a successful candidate.\xa0 In determining this range, we consider the experience, level of education (if applicable to the role), knowledge, skills, and abilities required to be had by a successful candidate as well as the budget and the company’s pay rates, generally. This said, we may have to make changes to our compensation estimates and job descriptions from time to time and we expressly reserve the right to do so.\xa0 Should we make any such changes, this advertisement will be revised to reflect such revisions.\xa0 We encourage you to occasionally re-visit this advertisement to ensure that you are abreast of any changes.\xa0 Thank you for your interest in joining Forbes!\nForbes aims to offer employees the flexibility they need in order to be successful. Some positions may require candidates to be based in a specific location for consideration while some roles may be fully remote (within the U.S.) if it aligns with the needs of the position. This position is only open to candidates residing in California, Colorado, Connecticut, District of Columbia, Florida, Georgia, Maine, Maryland, Massachusetts, New Jersey, New York, North Carolina, Pennsylvania, South Carolina, Tennessee, Texas, & Washington. Due to business operations and compliance requirements, we are unable to consider applicants based outside these states at this time.\nCreate a Job Alert\nInterested in building your career at Forbes? Get future opportunities sent straight to your email.\nAccepted file types: pdf, doc, docx, txt, rtf Forbes is an iconic global media brand that has symbolized success for over a century. Fueled by journalism that informs and inspires, Forbes spotlights the doers and doings shaping industries, achieving success and making an impact on the world. Forbes connects and convenes the most influential communities ranging from billionaires, business leaders and rising entrepreneurs to creators and innovators. The Forbes brand reaches more than 140 million people monthly worldwide through its trusted journalism, signature ForbesLive events and 49 licensed local editions in 81 countries.\nForbes Media is hiring a Manager, Social Sales & Newsletters to join our AI & Strategic Platforms team. This role will focus on driving revenue, strategic business development and representing the social and newsletter products at Forbes. This includes collaborating with key stakeholders at Forbes to drive sales, develop creative strategy for social activations, create new products and work with clients on bringing social advertising to life. This role will report into the Senior Director, Client Activation & Paid Social.\nOur office operates as a remote and hybrid workforce, with in-person collaboration expected for certain roles. Based on the responsibilities of this position, a hybrid work schedule with a minimum of one in-person collaboration day per week at our Jersey City, NJ headquarters is expected.\nParticipating in client-facing conversations as a social media and newsletter expert on behalf of Forbes when needed\nBrainstorm and ideate social & newsletter packages attached to Forbes’ products including editorial, live partnerships, content and social only\nBrainstorm and bring to market new social activations and opportunities for Forbes Live Event advertisers\nLiaison with key Forbes stakeholders including sales reps, account managers, sales planners, editorial social, live programming, integrated marketing, video producers, etc.\nPartner with external partners including LinkedIn, Meta, TikTok, influencer talent agencies, etc.\nForbes is committed to diversity, equity and inclusion and uses our various platforms to help the world build back with equity at its core. As the leading business media brand, we have a responsibility to ensure our content and experiences reflect the diverse audiences we serve. We have the unique ability to do good, leveraging our considerable editorial voice to highlight inequities and challenge systems.',
    'Firmware Engineer  Evaluate, benchmark, and select SoC/MCU platforms suitable for vision workloads and potential cloud integration.\nPrototype and deploy vision pipelines: capture raw image frames, perform preprocessing (ISP, color conversions), and manage AI inference models either locally or via cloud services.\nCollaborate with hardware teams to select optimal image sensors, lenses, and camera modules for object detection and classification tasks.\nImplement low-level drivers (I²C/SPI, GPIO, PWM) and incorporate them into device trees and board-support packages.\nWork with other team members to define and implement inter-processor communication protocols (SPI/UART/I²C) for sensor data exchange and system telemetry.\nEnsure coexistence of camera subsystems with other embedded peripherals (touchscreen, motors, sensors, connectivity modules) without performance degradation.\nExperience with OTA firmware updates for MCUs (bootloaders) and SoCs (partition management). Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, or related discipline, or equivalent experience.\n5+ years experience in embedded firmware engineering, including at least 3 years focused on camera or vision applications.\nStrong proficiency in C/C++ (Linux kernel drivers or RTOS firmware), with knowledge of memory management, DMA, interrupts, and real-time system constraints.\nHands-on experience with camera interfaces: MIPI-CSI-2, USB Video Class (UVC), or parallel RGB; comfortable working with image sensor datasheets, register programming, and sensor configuration.\nFamiliarity with vision frameworks (OpenCV, TensorRT, TensorFlow Lite, Edge TPU) or cloud-based AI processing platforms (AWS, Azure, Google Cloud); capable of prototyping object detection or segmentation pipelines, profiling performance, and suggesting optimizations.\nSolid understanding of I²C, SPI, UART, and GPIO interfaces.\nProven prototyping ability—quickly translating high-level performance targets into functional prototypes and effectively evaluating hardware-software tradeoffs.\nSkilled in using logic analyzers, oscilloscopes, and debugging tools (JTAG, GDB) to troubleshoot camera issues.\nExcellent communication skills, capable of producing clear documentation, architecture diagrams, and engaging in effective cross-team collaboration.\nExperience with camera calibration techniques, including lens distortion correction and color-space conversions.\nExperience with integrating touchscreen panels (LVDS/DSI/HDMI) and touch controllers (I²C/SPI), and developing responsive user interfaces.\nDevelop and optimize embedded UI frameworks (e.g., Qt Embedded, LVGL).\nFamiliarity with Linux system architectures, cross-compilation, build systems, and kernel module integration.\nPrevious experience in startups or rapid prototyping environments.\nFamiliarity with Agile methodologies, version control (GitHub), CI/CD pipelines, and project management tools (JIRA).\nInterested in building your career at Mill? Get future opportunities sent straight to your email.\nAccepted file types: pdf, doc, docx, txt, rtf Mill is all about answering a simple question: how can we prevent waste? Less waste can save time, money, energy, maybe even our planet. And there’s no better place to start than food. Food waste is one of the most solvable climate problems facing us today. Plus, our trash really stinks. It’s gross, heavy, and our least favorite chore. At Mill we are striving to build a better environment for all, as we take on climate and kitchen change.\nWe are seeking an extraordinary Firmware Engineer to lead the vision system development for our next-generation product. You will evaluate SoC and MCU platforms, and develop camera-based AI applications for object detection (either on-device or cloud-based). Collaborating closely with hardware designers, connectivity experts, and system engineers, your work will be crucial in turning innovative concepts into functional prototypes and products.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[     1.0000,      0.7013,     -0.0001],
#         [     0.7013,      1.0000,      0.1803],
#         [    -0.0001,      0.1803,      1.0000]])

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

Unnamed Dataset

• Size: 391 training samples
• Columns: <code>sentence_0</code> and <code>sentence_1</code>
• Approximate statistics based on the first 391 samples:

  	sentence_0	sentence_1
  type	string	string
  details	<ul><li>min: 2 tokens</li><li>mean: 7.03 tokens</li><li>max: 30 tokens</li></ul>	<ul><li>min: 2 tokens</li><li>mean: 240.47 tokens</li><li>max: 256 tokens</li></ul>

• Samples:

  sentence_0	sentence_1
  <code>Electronics Engineer</code>	<code>Electronics Engineer National Telecommunications and Information Administration Perform basic or applied research or engineering in such areas as propagation theory, scattering and diffraction, antenna design and optimization, radiated electromagnetic interference, and microwave systems.<br>Oversee and conduct lab and field measurements and/or develop, refine, and validate state-of-the-art theoretical and empirical radio propagation, spectrum, or wireless system models.<br>Develop software tools that use measurement findings and simulations to evaluate wireless system performance, predict technology behavior under various deployment conditions, and recommend more efficient uses of the radio spectrum.<br>Apply digital signal processing methods-including techniques for noise reduction, sampling, spectral estimation, filtering, and real-time detection and estimation-in software and hardware to analyze wireless signals.<br>Develop research proposals and present plans, progress, financials, and finding...</code>
  <code>Procurement Agent</code>	<code>Procurement Agent Boeing Communicate performance expectations and metrics to the supplier<br>Travel to suppliers and conduct assessments to ensure the supplier’s ongoing capability to support Boeing<br>Provide oversight and management of the supply and demand of parts, schedules, and supplier quality and delivery<br>Analyze and track Key Performance Indicators (KPIs) (things like on-time delivery) and using those indicators to predict risks; assist suppliers to develop new predictive indicators as needed<br>Mitigate discovered risks and initiate and implement supplier recovery plans and improvement activities<br>Lead Root Cause/Corrective Action initiatives to improve systemic issues<br>Facilitate cross-functional discussions regarding technical approaches, the resolution of quality issues, and the improvement of products and processes<br>Perform some level of system transactions and data entry to maintain Enterprise Resource Planning (ERP) and/or SAP fidelity, support metric generation, and maintain stake...</code>
  <code>Digital Design Developer, Opinion</code>	<code>Digital Design Developer, Opinion The New York Times Company Office: New York, NY Department: Editing<br>Develop templates for recurring editorial features.<br>Optimize projects to load and perform across browsers and devices.<br>You will report to Deputy Editor Design.<br>Familiarity with Git and collaborative coding workflows 5+ years of digital design experience<br>Visual and interactive design skills — a clear grasp of typography, layout, color, motion, and behavior<br>Experience in client-side Web development, including JavaScript, HTML and CSS required<br>Experience working with video, audio, graphics<br>Experience or familiarity with Svelte<br>Ability to work in a fast-paced news environment with a willingness to work a flexible, news-driven schedule that may occasionally include nights or weekends The mission of The New York Times is to seek the truth and help people understand the world. That means independent journalism is at the heart of all we do as a company. It’s why we have a world-renowned news...</code>

• Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 1
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

<details><summary>Click to expand</summary>

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• 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
• num_train_epochs: 1
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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: 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
• 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: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Framework Versions

• Python: 3.12.11
• Sentence Transformers: 5.1.1
• Transformers: 4.56.2
• PyTorch: 2.8.0+cu126
• Accelerate: 1.10.1
• Datasets: 4.0.0
• 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",
}

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}
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->