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
Maximum Sequence Length: 256 tokens
Output Dimensionality: 384 tokens
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': 256, 'do_lower_case': False}) with Transformer model: 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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    'Experience in building materials industry',
    'As a software developer, I have worked on various applications that streamline project management processes. My experience includes developing user-friendly interfaces and integrating third-party APIs to enhance functionality. I have collaborated with cross-functional teams to gather requirements and deliver solutions that meet client needs. My role involved conducting code reviews and ensuring adherence to best practices in software development. I have a strong understanding of programming languages such as JavaScript, Python, and Ruby, which I have utilized to build robust applications. Additionally, I have experience in database management and optimization, ensuring that data is stored efficiently. My background includes working in Agile environments, where I have participated in sprint planning and retrospectives. I have also contributed to documentation efforts to ensure that all project details are captured accurately. Overall, my software development experience has equipped me with the skills necessary to create effective solutions for various industries.',
    'In my role as a quality assurance engineer, I was responsible for testing software applications to ensure they met quality standards. I collaborated with development teams to understand application functionality and create test plans. My work involved executing test cases and documenting results to identify defects. I also implemented automated testing strategies to improve efficiency in the testing process. My experience with various testing tools allowed me to streamline testing efforts and ensure comprehensive coverage. I regularly participated in team meetings to discuss testing progress and address any issues that arose. Additionally, I provided feedback to developers to help improve application quality. My strong attention to detail helped in identifying potential issues before deployment. I also contributed to documentation that outlined testing processes and best practices.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

Dataset: validation
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.9471
spearman_cosine	0.9124
pearson_manhattan	0.9358
spearman_manhattan	0.9118
pearson_euclidean	0.9344
spearman_euclidean	0.9124
pearson_dot	0.9471
spearman_dot	0.9124
pearson_max	0.9471
spearman_max	0.9124

Training Details

Training Dataset

Unnamed Dataset

Size: 375,259 training samples
Columns: text1, text2, and label
Approximate statistics based on the first 1000 samples:
text1 text2 label
type string string float
details
min: 6 tokens
mean: 13.95 tokens
max: 22 tokens
min: 7 tokens
mean: 152.64 tokens
max: 224 tokens
min: 0.0
mean: 0.48
max: 1.0

Samples:

text1	text2	label
`Experience with APIs`	In my previous role, I was responsible for developing and maintaining a robust e-commerce platform that integrated various third-party services. This involved creating seamless connections between our application and external payment gateways, ensuring secure transactions. I utilized RESTful services to facilitate communication between the front-end and back-end systems, which significantly improved the user experience. Additionally, I implemented OAuth for secure authentication, allowing users to log in using their social media accounts. My work also included writing comprehensive documentation for the API endpoints, which helped other developers understand how to interact with our services. I collaborated closely with the QA team to ensure that all API functionalities were thoroughly tested before deployment. My experience with APIs has equipped me with the skills to troubleshoot and optimize performance issues effectively. I also participated in code reviews, providing feedback on best practices for API development. Overall, my background in API integration has been crucial in delivering high-quality software solutions.	`1.0`
`Experience with APIs`	As a software engineer, I played a key role in a project that involved building a mobile application for a healthcare provider. My responsibilities included designing the architecture of the application, which required extensive use of APIs to connect with various health data sources. I ensured that the application could retrieve patient information securely and efficiently through well-defined API calls. My experience with APIs allowed me to implement error handling and data validation mechanisms, which enhanced the application's reliability. I also worked on creating a user-friendly interface that displayed data fetched from the APIs in a meaningful way. Collaborating with cross-functional teams, I gathered requirements and translated them into technical specifications for the API development. I conducted performance testing to ensure that the API responses were quick and met user expectations. Additionally, I provided training sessions for junior developers on best practices for API usage and integration. This experience has solidified my understanding of API-driven development.	`1.0`
`Experience with APIs`	In my role as a backend developer, I was tasked with creating a microservices architecture for a large-scale application. This involved developing multiple APIs that allowed different services to communicate with each other efficiently. I focused on ensuring that the APIs were RESTful and followed industry standards for data exchange. My experience with APIs included implementing versioning strategies to maintain backward compatibility as the application evolved. I also utilized tools like Postman for testing the APIs and ensuring they met the required specifications. My work required close collaboration with front-end developers to ensure that the APIs provided the necessary data for the user interface. I also monitored API performance and made optimizations to reduce latency and improve response times. Additionally, I documented the API endpoints thoroughly, which facilitated easier onboarding for new team members. This experience has been invaluable in my career as it has deepened my understanding of service-oriented architecture.	`1.0`

Loss: CosineSimilarityLoss with these parameters:
json
1{ 2 "loss_fct": "torch.nn.modules.loss.MSELoss" 3}

Evaluation Dataset

Unnamed Dataset

Size: 20,089 evaluation samples
Columns: text1, text2, and label
Approximate statistics based on the first 1000 samples:
text1 text2 label
type string string float
details
min: 7 tokens
mean: 14.41 tokens
max: 32 tokens
min: 7 tokens
mean: 146.8 tokens
max: 226 tokens
min: 0.0
mean: 0.48
max: 1.0

Samples:

text1	text2	label
`Experience with API security protocols such as OAuth2 or JWT`	In my previous role as a software engineer, I was responsible for developing secure web applications that required robust authentication mechanisms. I implemented various security measures, including encryption and secure data transmission. My work involved integrating third-party services, which necessitated a deep understanding of API security protocols. I successfully utilized OAuth2 for user authentication, ensuring that sensitive data was protected during transactions. Additionally, I collaborated with the security team to conduct regular audits and vulnerability assessments. My experience also includes designing RESTful APIs that adhere to best practices in security. I have a strong foundation in web application security principles, which I applied to enhance the overall security posture of our applications. My ability to communicate complex security concepts to non-technical stakeholders was crucial in fostering a security-aware culture within the team.	`1.0`
`Experience with API security protocols such as OAuth2 or JWT`	As a backend developer, I focused on creating scalable microservices that interacted with various external APIs. A key aspect of my role involved ensuring that these services were secure and compliant with industry standards. I implemented token-based authentication using JWT, which streamlined the user experience while maintaining high security. My responsibilities included writing comprehensive documentation for API endpoints, detailing the security measures in place. I also participated in code reviews, where I emphasized the importance of secure coding practices. Collaborating with cross-functional teams, I contributed to the development of a security framework that guided our API development. My experience in this area has equipped me with the skills to identify potential security risks and implement effective mitigation strategies. I am passionate about staying updated on the latest security trends and best practices.	`1.0`
`Experience with API security protocols such as OAuth2 or JWT`	In my role as a systems architect, I was tasked with designing a secure infrastructure for our cloud-based applications. This involved selecting appropriate security protocols to protect user data and ensure compliance with regulations. I chose to implement OAuth2 for authorization, which allowed for secure access to our APIs. My work included developing a comprehensive security policy that outlined the use of various protocols and best practices. I also led training sessions for the development team on the importance of API security and how to implement these protocols effectively. My experience in threat modeling helped identify potential vulnerabilities in our architecture. I collaborated closely with the DevOps team to automate security checks in our CI/CD pipeline. This proactive approach significantly reduced the risk of security breaches in our applications.	`1.0`

Loss: CosineSimilarityLoss with these parameters:
json
1{ 2 "loss_fct": "torch.nn.modules.loss.MSELoss" 3}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 128
per_device_eval_batch_size: 128
num_train_epochs: 10.0
warmup_ratio: 0.1
fp16: True
load_best_model_at_end: True
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 128
per_device_eval_batch_size: 128
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: 10.0
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
use_ipex: False
bf16: False
fp16: True
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: True
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
dispatch_batches: None
split_batches: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
eval_use_gather_object: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	validation_spearman_max
0.1705	500	0.0737	-	-
0.3411	1000	0.0407	0.0377	0.8736
0.5116	1500	0.0346	-	-
0.6821	2000	0.0289	0.0260	0.8932
0.8527	2500	0.0275	-	-
1.0201	3000	0.0279	0.0317	0.8718
1.1907	3500	0.026	-	-
1.3612	4000	0.0241	0.0233	0.9000
1.5317	4500	0.0231	-	-
1.7023	5000	0.0203	0.0206	0.9045
1.8728	5500	0.0197	-	-
2.0402	6000	0.0212	0.0218	0.9018
2.2108	6500	0.0187	-	-
2.3813	7000	0.0192	0.0188	0.9062
2.5518	7500	0.0185	-	-
2.7224	8000	0.0163	0.0185	0.9083
2.8929	8500	0.0164	-	-
3.0604	9000	0.0169	0.0205	0.9075
3.2309	9500	0.0156	-	-
3.4014	10000	0.0161	0.0179	0.9087
3.5720	10500	0.016	-	-
3.7425	11000	0.0139	0.0179	0.9101
3.9130	11500	0.0143	-	-
4.0805	12000	0.0143	0.0190	0.9104
4.2510	12500	0.0136	-	-
4.4216	13000	0.0144	0.0212	0.9093
4.5921	13500	0.0139	-	-
4.7626	14000	0.0123	0.0174	0.9116
4.9332	14500	0.0126	-	-
5.1006	15000	0.0129	0.0181	0.9109
5.2711	15500	0.0122	-	-
5.4417	16000	0.0128	0.0210	0.9100
5.6122	16500	0.0123	-	-
5.7827	17000	0.011	0.0178	0.9116
5.9533	17500	0.0114	-	-
6.1207	18000	0.0115	0.0175	0.9124

Framework Versions

Python: 3.12.3
Sentence Transformers: 3.0.1
Transformers: 4.44.2
PyTorch: 2.4.0+cu121
Accelerate: 0.33.0
Datasets: 2.21.0
Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

bibtex

@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",
}

	text1	text2	label
type	string	string	float
details	min: 6 tokens mean: 13.95 tokens max: 22 tokens	min: 7 tokens mean: 152.64 tokens max: 224 tokens	min: 0.0 mean: 0.48 max: 1.0

recruitco/embedding_criteria_profile_summary_matching_from_criteria_minilm_v3

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

Model Details

Model Description

Model Sources

Full Model Architecture

Usage

Direct Usage (Sentence Transformers)

Evaluation

Metrics

Semantic Similarity

Training Details

Training Dataset

Unnamed Dataset

Evaluation Dataset

Unnamed Dataset

Training Hyperparameters

Non-Default Hyperparameters

All Hyperparameters

Training Logs

Framework Versions

Citation

BibTeX

Sentence Transformers

recruitco/
embedding_criteria_profile_summary_matching_from_criteria_minilm_v3