SentenceTransformer based on newmindai/TurkEmbed4STS

This is a sentence-transformers model finetuned from newmindai/TurkEmbed4STS. 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: newmindai/TurkEmbed4STS
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 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': 512, 'do_lower_case': False}) with Transformer model: NewModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, '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': 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("sentence_transformers_model_id")
# Run inference
sentences = [
    'Bir yıl tam yelken üniversitesini ne kadar ödeyeceğim',
    "Full Sail Üniversitesi'nde öğrenim ve ücretler 23,117 $ mali yardım olmadan. Oda, yönetim kurulu ve diğer ücretlerin bir araya getirilmesiyle, toplam katılım maliyeti 23,647 $ 'dır. Maliyet ve Yardım. Pahalı öğrenim (23.117 $, eyalet içi ve eyalet dışı öğrenciler için aynı)",
    'Hangi hemoglobin seviyesi anemik olarak kabul edilir - ben anemikim ve 88 g / l hemoglobin seviyesine sahipim. Bu ne kadar ciddi? 88 mi demek istediniz? Normal hemoglobin yaklaşık 15 (+/- az) bu yüzden 8 civarında gerçekten çok düşüktür. Bu aneminin nedeni araştırılmış ve düzgün bir şekilde tedavi edilmiş olmalıdır. Bu arada kan transfüzyonuna ihtiyacınız olabilir',
]
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]

Training Details

Training Dataset

Unnamed Dataset

Size: 1,840,212 training samples
Columns: sentence_0, sentence_1, and label
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 label
type string string float
details
min: 4 tokens
mean: 10.24 tokens
max: 26 tokens

min: 16 tokens
mean: 79.54 tokens
max: 138 tokens

min: 0.0
mean: 0.49
max: 1.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 4 tokens mean: 10.24 tokens max: 26 tokens	min: 16 tokens mean: 79.54 tokens max: 138 tokens	min: 0.0 mean: 0.49 max: 1.0

Samples:

sentence_0	sentence_1	label
`Hipotez ve teori tanımı`	Teori ve Uygulamada Tanım Hakkında. Tanım sorunu uzun bir geçmişe sahiptir ve Platon ve Aristoteles'ten başlayarak Batı geleneğinin en seçkin düşünürlerinden bazılarının zihinlerini meşgul etmiştir. Ancak aynı zamanda modern toplumun bağlı olduğu sayısız metinleri hazırlamak veya yorumlamak zorunda olan herkesin sürekli karşılaştığı günlük bir sorundur. Teori ve Uygulamada Tanım, zorlukların özellikle akut bir biçimde ortaya çıktığı iki alana odaklanmaktadır: lexicography ve hukuk.	`0.0`
`Oturma ücreti nedir`	`Devamını Oku 1 TÜM PORTRAİT SESSIONS INCLUDE... 2 Oturum Ücreti fotoğrafçının Zaman ve Yeteneklerini kapsar. 3 Unutma... Vergiyi ekleyin! (%8.4 vergi ödemeniz gerekmektedir.) Portre seansınızdan önce oturma ücreti alınır.`	`1.0`
`Kafeine ne kadar l-theanine`	`Kafeini düşünün. Espresso ve kahvenin her ikisi de doğal olarak siyah ve yeşil çaylarda olduğu gibi kafeine sahiptir. Eğer içeceğinizde daha az kafein istiyorsanız, yarım kafein (1/2 normal kafein miktarı) veya kafeinsiz (kafeinsiz) sipariş edin. Ayrıca, içeceğinizden biraz daha fazla enerji arıyorsanız, ekstra kahve de ekleyebilirsiniz.`	`0.0`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "CachedMultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 1024
per_device_eval_batch_size: 1024
num_train_epochs: 1
fp16: True
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 1024
per_device_eval_batch_size: 1024
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: 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: 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}
tp_size: 0
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: None
hub_always_push: False
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
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss
0.2781	500	9.66
0.5562	1000	8.0512
0.8343	1500	7.7153

Framework Versions

Python: 3.11.12
Sentence Transformers: 4.1.0
Transformers: 4.51.3
PyTorch: 2.6.0+cu124
Accelerate: 1.6.0
Datasets: 3.6.0
Tokenizers: 0.21.1

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}
}

CachedMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

ozayezerceli
/

TurkEmded4Retrievalv2