metadata

language: []
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:4068
  - loss:MultipleNegativesRankingLoss
base_model: distilbert/distilroberta-base
datasets: []
metrics:
  - cosine_accuracy
  - dot_accuracy
  - manhattan_accuracy
  - euclidean_accuracy
  - max_accuracy
widget:
  - source_sentence: >-
      Proficiency in C# scripting is essential for creating custom scripts and
      extensions to enhance ABBYY FlexiCapture and ABBYY Vantage functionality.
    sentences:
      - Successfully presented financial reports to executives
      - Worked on improving user interfaces using HTML and CSS
      - Created extensions to optimize data capture processes
  - source_sentence: >-
      Knowledgeable in supporting Cyber Security Operations and investigation
      requests.
    sentences:
      - Assisted in incident response for security breaches
      - Coordinated communication strategies for corporate events
      - Developed mobile applications for e-commerce
  - source_sentence: >-
      Bachelor’s degree in Human Resources, Business Administration, Finance or
      related field
    sentences:
      - prepared monthly production reports for management meetings
      - Bachelor of Science in Human Resources Management
      - Completed a course in Marketing Strategy
  - source_sentence: >-
      A strong interest in photography or videography is necessary for this
      role.
    sentences:
      - produced short promotional videos for social media platforms
      - Conducted training sessions for new software implementations
      - conducted market research on competitor strategies
  - source_sentence: Ability to work both independently and as part of a collaborative team.
    sentences:
      - Worked in isolation and avoided team interactions
      - Participated in team meetings and contributed to group problem-solving
      - Authored clear documentation for complex data processes
pipeline_tag: sentence-similarity
model-index:
  - name: SentenceTransformer based on distilbert/distilroberta-base
    results:
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: all nli dev
          type: all-nli-dev
        metrics:
          - type: cosine_accuracy
            value: 0.9803536345776032
            name: Cosine Accuracy
          - type: dot_accuracy
            value: 0.047151277013752456
            name: Dot Accuracy
          - type: manhattan_accuracy
            value: 0.9724950884086444
            name: Manhattan Accuracy
          - type: euclidean_accuracy
            value: 0.9724950884086444
            name: Euclidean Accuracy
          - type: max_accuracy
            value: 0.9803536345776032
            name: Max Accuracy
          - type: cosine_accuracy
            value: 0.9803536345776032
            name: Cosine Accuracy
          - type: dot_accuracy
            value: 0.047151277013752456
            name: Dot Accuracy
          - type: manhattan_accuracy
            value: 0.9724950884086444
            name: Manhattan Accuracy
          - type: euclidean_accuracy
            value: 0.9724950884086444
            name: Euclidean Accuracy
          - type: max_accuracy
            value: 0.9803536345776032
            name: Max Accuracy

SentenceTransformer based on distilbert/distilroberta-base

This is a sentence-transformers model finetuned from distilbert/distilroberta-base. 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: distilbert/distilroberta-base
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 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': 512, 'do_lower_case': False}) with Transformer model: RobertaModel 
  (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("trbeers/distilroberta-base-nli-v0.2")
# Run inference
sentences = [
    'Ability to work both independently and as part of a collaborative team.',
    'Participated in team meetings and contributed to group problem-solving',
    'Worked in isolation and avoided team interactions',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

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

Evaluation

Metrics

Triplet

Dataset: all-nli-dev
Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.9804
dot_accuracy	0.0472
manhattan_accuracy	0.9725
euclidean_accuracy	0.9725
max_accuracy	0.9804

Triplet

Dataset: all-nli-dev
Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.9804
dot_accuracy	0.0472
manhattan_accuracy	0.9725
euclidean_accuracy	0.9725
max_accuracy	0.9804

Training Details

Training Dataset

Unnamed Dataset

Size: 4,068 training samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 8 tokens mean: 16.67 tokens max: 37 tokens	min: 7 tokens mean: 11.82 tokens max: 22 tokens	min: 5 tokens mean: 9.13 tokens max: 15 tokens

Samples:

anchor	positive	negative
`Experience in managing meetings with program participants and tracking action items effectively.`	`Coordinated project meetings and followed up on team tasks`	`Assisted in developing marketing strategies`
`Ability to replace faulty electrical components with precision.`	`Conducted detailed inspections of wiring and circuits`	`Handled plumbing repairs and maintenance tasks`
`Knowledge of loss prevention, security, and safety protocols.`	`Implemented safety measures in warehouse operations`	`Worked as a sales associate`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

Unnamed Dataset

Size: 1,018 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: 16.56 tokens max: 42 tokens	min: 6 tokens mean: 11.77 tokens max: 20 tokens	min: 5 tokens mean: 9.0 tokens max: 17 tokens

Samples:

anchor	positive	negative
`The ability to complete a background investigation and drug screen is necessary for employment.`	`Conducted thorough background investigations for security personnel`	`Managed scheduling for office staff`
`Ability to create compelling business cases to drive organizational change.`	`Developed comprehensive business cases that successfully led to strategic organizational changes`	`Managed project timelines and budgets for software development projects`
`Proven understanding of ERP concepts and their applications in business.`	`Conducted workshops on business process improvement`	`Managed social media accounts`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 128
per_device_eval_batch_size: 128
num_train_epochs: 1
warmup_ratio: 0.1
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
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: 1
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: 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}
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
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	loss	all-nli-dev_max_accuracy
0	0	-	0.9607
0.3125	10	2.0385	0.9587
0.625	20	1.5189	0.9745
0.9375	30	1.3685	0.9804
1.0	32	-	0.9804

Framework Versions

Python: 3.10.11
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.1
Accelerate: 0.31.0
Datasets: 2.19.1
Tokenizers: 0.19.1

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}
}