metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:5749
  - loss:CosineSimilarityLoss
base_model: CocoRoF/ModernBERT-SimCSE_v02
widget:
  - source_sentence: 우리는 움직이는 동행 우주 정지 좌표계에 비례하여 이동하고 있습니다 ... 약 371km / s에서 별자리 leo 쪽으로. "
    sentences:
      - 두 마리의 독수리가 가지에 앉는다.
      - 다른 물체와는 관련이 없는 '정지'는 없다.
      - 소녀는 버스의 열린 문 앞에 서 있다.
  - source_sentence: 숲에는 개들이 있다.
    sentences:
      - 양을 보는 아이들.
      - 여왕의 배우자를 "왕"이라고 부르지 않는 것은 아주 좋은 이유가 있다. 왜냐하면 그들은 왕이 아니기 때문이다.
      - 개들은 숲속에 혼자 있다.
  - source_sentence: '첫째, 두 가지 다른 종류의 대시가 있다는 것을 알아야 합니다 : en 대시와 em 대시.'
    sentences:
      - 그들은 그 물건들을 집 주변에 두고 가거나 집의 정리를 해칠 의도가 없다.
      - 세미콜론은 혼자 있을 수 있는 문장에 참여하는데 사용되지만, 그들의 관계를 강조하기 위해 결합됩니다.
      - 그의 남동생이 지켜보는 동안 집 앞에서 트럼펫을 연주하는 금발의 아이.
  - source_sentence: 한 여성이 생선 껍질을 벗기고 있다.
    sentences:
      - 한 남자가 수영장으로 뛰어들었다.
      - 한 여성이 프라이팬에 노란 혼합물을 부어 넣고 있다.
      - 두 마리의 갈색 개가 눈 속에서 서로 놀고 있다.
  - source_sentence: 버스가 바쁜 길을 따라 운전한다.
    sentences:
      - 우리와 같은 태양계가 은하계 밖에서 존재할 수도 있을 것입니다.
      - 그 여자는 데이트하러 가는 중이다.
      - 녹색 버스가 도로를 따라 내려간다.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_euclidean
  - spearman_euclidean
  - pearson_manhattan
  - spearman_manhattan
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
model-index:
  - name: SentenceTransformer based on CocoRoF/ModernBERT-SimCSE_v02
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts dev
          type: sts_dev
        metrics:
          - type: pearson_cosine
            value: 0.8223949445074785
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8220107207834706
            name: Spearman Cosine
          - type: pearson_euclidean
            value: 0.7785831525283676
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.7815628643916452
            name: Spearman Euclidean
          - type: pearson_manhattan
            value: 0.7809119630672191
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.7846536514745763
            name: Spearman Manhattan
          - type: pearson_dot
            value: 0.7543765794886113
            name: Pearson Dot
          - type: spearman_dot
            value: 0.7434525191412167
            name: Spearman Dot
          - type: pearson_max
            value: 0.8223949445074785
            name: Pearson Max
          - type: spearman_max
            value: 0.8220107207834706
            name: Spearman Max

SentenceTransformer based on CocoRoF/ModernBERT-SimCSE_v02

This is a sentence-transformers model finetuned from CocoRoF/ModernBERT-SimCSE_v02. 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: CocoRoF/ModernBERT-SimCSE_v02
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: ModernBertModel 
  (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})
  (2): Dense({'in_features': 768, 'out_features': 768, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
)

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("CocoRoF/ModernBERT-SimCSE-multitask_v03")
# Run inference
sentences = [
    '버스가 바쁜 길을 따라 운전한다.',
    '녹색 버스가 도로를 따라 내려간다.',
    '그 여자는 데이트하러 가는 중이다.',
]
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

Semantic Similarity

Dataset: sts_dev
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8224
spearman_cosine	0.822
pearson_euclidean	0.7786
spearman_euclidean	0.7816
pearson_manhattan	0.7809
spearman_manhattan	0.7847
pearson_dot	0.7544
spearman_dot	0.7435
pearson_max	0.8224
spearman_max	0.822

Training Details

Training Dataset

Unnamed Dataset

Size: 5,749 training samples
Columns: sentence1, sentence2, and score
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 score
type string string float
details
min: 7 tokens
mean: 13.52 tokens
max: 36 tokens

min: 7 tokens
mean: 13.41 tokens
max: 32 tokens

min: 0.0
mean: 0.45
max: 1.0

	sentence1	sentence2	score
type	string	string	float
details	min: 7 tokens mean: 13.52 tokens max: 36 tokens	min: 7 tokens mean: 13.41 tokens max: 32 tokens	min: 0.0 mean: 0.45 max: 1.0

Samples:

sentence1	sentence2	score
`비행기가 이륙하고 있다.`	`비행기가 이륙하고 있다.`	`1.0`
`한 남자가 큰 플루트를 연주하고 있다.`	`남자가 플루트를 연주하고 있다.`	`0.76`
`한 남자가 피자에 치즈를 뿌려놓고 있다.`	`한 남자가 구운 피자에 치즈 조각을 뿌려놓고 있다.`	`0.76`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Evaluation Dataset

Unnamed Dataset

Size: 1,500 evaluation samples
Columns: sentence1, sentence2, and score
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 score
type string string float
details
min: 7 tokens
mean: 20.38 tokens
max: 52 tokens

min: 6 tokens
mean: 20.52 tokens
max: 54 tokens

min: 0.0
mean: 0.42
max: 1.0

	sentence1	sentence2	score
type	string	string	float
details	min: 7 tokens mean: 20.38 tokens max: 52 tokens	min: 6 tokens mean: 20.52 tokens max: 54 tokens	min: 0.0 mean: 0.42 max: 1.0

Samples:

sentence1	sentence2	score
`안전모를 가진 한 남자가 춤을 추고 있다.`	`안전모를 쓴 한 남자가 춤을 추고 있다.`	`1.0`
`어린아이가 말을 타고 있다.`	`아이가 말을 타고 있다.`	`0.95`
`한 남자가 뱀에게 쥐를 먹이고 있다.`	`남자가 뱀에게 쥐를 먹이고 있다.`	`1.0`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

overwrite_output_dir: True
eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
gradient_accumulation_steps: 8
learning_rate: 1e-05
num_train_epochs: 10.0
warmup_ratio: 0.1
push_to_hub: True
hub_model_id: CocoRoF/ModernBERT-SimCSE-multitask_v03
hub_strategy: checkpoint
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: True
do_predict: False
eval_strategy: steps
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: 8
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 1e-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: 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: True
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: True
resume_from_checkpoint: None
hub_model_id: CocoRoF/ModernBERT-SimCSE-multitask_v03
hub_strategy: checkpoint
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
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
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	sts_dev_spearman_max
0.2228	10	0.0283	-	-
0.4457	20	0.0344	-	-
0.6685	30	0.0305	0.0310	0.7939
0.8914	40	0.0489	-	-
1.1337	50	0.0382	-	-
1.3565	60	0.0271	0.0293	0.7994
1.5794	70	0.0344	-	-
1.8022	80	0.0382	-	-
2.0446	90	0.0419	0.0280	0.8059
2.2674	100	0.0244	-	-
2.4903	110	0.0307	-	-
2.7131	120	0.0291	0.0269	0.8108
2.9359	130	0.038	-	-
3.1783	140	0.0269	-	-
3.4011	150	0.0268	0.0262	0.8155
3.6240	160	0.0246	-	-
3.8468	170	0.0313	-	-
4.0891	180	0.0303	0.0259	0.8185
4.3120	190	0.0198	-	-
4.5348	200	0.0257	-	-
4.7577	210	0.0242	0.0255	0.8202
4.9805	220	0.0293	-	-
5.2228	230	0.0193	-	-
5.4457	240	0.0222	0.0254	0.8222
5.6685	250	0.0184	-	-
5.8914	260	0.0243	-	-
6.1337	270	0.0204	0.0254	0.8235
6.3565	280	0.0147	-	-
6.5794	290	0.0196	-	-
6.8022	300	0.0176	0.0253	0.8227
7.0446	310	0.0202	-	-
7.2674	320	0.0123	-	-
7.4903	330	0.0151	0.0254	0.8236
7.7131	340	0.0132	-	-
7.9359	350	0.0158	-	-
8.1783	360	0.0118	0.0256	0.8240
8.4011	370	0.0115	-	-
8.6240	380	0.0105	-	-
8.8468	390	0.0111	0.0256	0.8215
9.0891	400	0.011	-	-
9.3120	410	0.0076	-	-
9.5348	420	0.0091	0.0256	0.8220
9.7577	430	0.0075	-	-
9.9805	440	0.0093	-	-

Framework Versions

Python: 3.11.10
Sentence Transformers: 3.3.1
Transformers: 4.48.0.dev0
PyTorch: 2.5.1+cu121
Accelerate: 1.1.0
Datasets: 3.1.0
Tokenizers: 0.21.0

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",
}