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 retrieval.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: sentence-transformers/all-MiniLM-L6-v2
Maximum Sequence Length: 256 tokens
Output Dimensionality: 384 dimensions
Similarity Function: Cosine Similarity
Supported Modality: Text

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'mean', '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("kelompoknlp2026dsindo/retriever_260511070554_FINAL")
# Run inference
sentences = [
    'Measurements indicating that 2017 had relatively more sea ice in the Arctic and less melting of glacial ice in Greenland casts scientific doubt on the reality of global warming.',
    'The effects of global warming in the Arctic, or climate change in the Arctic include rising air and water temperatures, loss of sea ice, and melting of the Greenland ice sheet with a related cold temperature anomaly, observed since the 1970s.',
    'Human-caused increases in greenhouse gases are responsible for most of the observed global average surface warming of roughly 0.8\u202f°C (1.5\u202f°F) over the past 140 years.',
]
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, 0.5034, 0.2555],
#         [0.5034, 1.0000, 0.4109],
#         [0.2555, 0.4109, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: dev
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.2389
cosine_accuracy@3	0.469
cosine_accuracy@5	0.5752
cosine_accuracy@10	0.6726
cosine_precision@1	0.2389
cosine_precision@3	0.1947
cosine_precision@5	0.1575
cosine_precision@10	0.1053
cosine_precision@50	0.0331
cosine_precision@100	0.0185
cosine_precision@200	0.0102
cosine_recall@1	0.1121
cosine_recall@3	0.2397
cosine_recall@5	0.3314
cosine_recall@10	0.4282
cosine_recall@50	0.6518
cosine_recall@100	0.7305
cosine_recall@200	0.8034
cosine_ndcg@10	0.324
cosine_mrr@10	0.3777
cosine_map@100	0.2577

Training Details

Training Dataset

Unnamed Dataset

Size: 2,192 training samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 7 tokens
mean: 26.25 tokens
max: 70 tokens
min: 9 tokens
mean: 37.84 tokens
max: 111 tokens

Samples:

anchor	positive
`Not only is there no scientific evidence that CO2 is a pollutant, higher CO2 concentrations actually help ecosystems support more plant and animal life.`	`At very high concentrations (100 times atmospheric concentration, or greater), carbon dioxide can be toxic to animal life, so raising the concentration to 10,000 ppm (1%) or higher for several hours will eliminate pests such as whiteflies and spider mites in a greenhouse.`
`Not only is there no scientific evidence that CO2 is a pollutant, higher CO2 concentrations actually help ecosystems support more plant and animal life.`	`Plants can grow as much as 50 percent faster in concentrations of 1,000 ppm CO 2 when compared with ambient conditions, though this assumes no change in climate and no limitation on other nutrients.`
`Not only is there no scientific evidence that CO2 is a pollutant, higher CO2 concentrations actually help ecosystems support more plant and animal life.`	`Higher carbon dioxide concentrations will favourably affect plant growth and demand for water.`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false, 5 "directions": [ 6 "query_to_doc" 7 ], 8 "partition_mode": "joint", 9 "hardness_mode": null, 10 "hardness_strength": 0.0 11}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 1e-05
num_train_epochs: 20
warmup_steps: 100
data_seed: 42
load_best_model_at_end: True
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

do_predict: False
prediction_loss_only: True
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
gradient_accumulation_steps: 1
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: 20
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_ratio: None
warmup_steps: 100
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
enable_jit_checkpoint: False
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
use_cpu: False
seed: 42
data_seed: 42
bf16: False
fp16: False
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: -1
ddp_backend: None
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: True
ignore_data_skip: False
fsdp: []
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
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
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
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
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_for_metrics: []
eval_do_concat_batches: True
auto_find_batch_size: False
full_determinism: False
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_num_input_tokens_seen: no
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: True
use_cache: False
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	dev_cosine_ndcg@10
0.0365	5	0.8437	-
0.0730	10	0.8164	-
0.1095	15	0.8009	-
0.1460	20	0.7637	-
0.1825	25	0.8168	-
0.2190	30	0.8028	-
0.2555	35	0.6182	-
0.2920	40	0.6689	-
0.3285	45	0.8032	-
0.3650	50	0.6298	-
0.4015	55	0.6443	-
0.4380	60	0.6813	-
0.4745	65	0.4928	-
0.5109	70	0.8207	-
0.5474	75	0.5841	-
0.5839	80	0.5238	-
0.6204	85	0.4142	-
0.6569	90	0.6077	-
0.6934	95	0.4521	-
0.7299	100	0.5201	-
0.7664	105	0.5142	-
0.8029	110	0.4498	-
0.8394	115	0.5953	-
0.8759	120	0.3598	-
0.9124	125	0.5734	-
0.9489	130	0.4794	-
0.9854	135	0.4036	-
1.0	137	-	0.3189
1.0219	140	0.3969	-
1.0584	145	0.4520	-
1.0949	150	0.4475	-
1.1314	155	0.5730	-
1.1679	160	0.5064	-
1.2044	165	0.3858	-
1.2409	170	0.4059	-
1.2774	175	0.1969	-
1.3139	180	0.6806	-
1.3504	185	0.4008	-
1.3869	190	0.3313	-
1.4234	195	0.5591	-
1.4599	200	0.3068	-
1.4964	205	0.2700	-
1.5328	210	0.3596	-
1.5693	215	0.4863	-
1.6058	220	0.3552	-
1.6423	225	0.3481	-
1.6788	230	0.3418	-
1.7153	235	0.5629	-
1.7518	240	0.4146	-
1.7883	245	0.5223	-
1.8248	250	0.4027	-
1.8613	255	0.3733	-
1.8978	260	0.3287	-
1.9343	265	0.3615	-
1.9708	270	0.2782	-
2.0	274	-	0.3254
2.0073	275	0.4479	-
2.0438	280	0.2815	-
2.0803	285	0.3969	-
2.1168	290	0.3915	-
2.1533	295	0.4366	-
2.1898	300	0.3599	-
2.2263	305	0.2776	-
2.2628	310	0.2626	-
2.2993	315	0.3017	-
2.3358	320	0.3362	-
2.3723	325	0.2068	-
2.4088	330	0.2746	-
2.4453	335	0.3807	-
2.4818	340	0.2927	-
2.5182	345	0.2699	-
2.5547	350	0.2439	-
2.5912	355	0.3890	-
2.6277	360	0.5438	-
2.6642	365	0.4341	-
2.7007	370	0.2560	-
2.7372	375	0.1545	-
2.7737	380	0.1854	-
2.8102	385	0.4939	-
2.8467	390	0.2932	-
2.8832	395	0.3529	-
2.9197	400	0.2541	-
2.9562	405	0.2895	-
2.9927	410	0.3107	-
3.0	411	-	0.3277
3.0292	415	0.3130	-
3.0657	420	0.2742	-
3.1022	425	0.2136	-
3.1387	430	0.1759	-
3.1752	435	0.2413	-
3.2117	440	0.2888	-
3.2482	445	0.1988	-
3.2847	450	0.2650	-
3.3212	455	0.1097	-
3.3577	460	0.2162	-
3.3942	465	0.2702	-
3.4307	470	0.2485	-
3.4672	475	0.2010	-
3.5036	480	0.2736	-
3.5401	485	0.2912	-
3.5766	490	0.3467	-
3.6131	495	0.2874	-
3.6496	500	0.2772	-
3.6861	505	0.3618	-
3.7226	510	0.1875	-
3.7591	515	0.2509	-
3.7956	520	0.2517	-
3.8321	525	0.2589	-
3.8686	530	0.3354	-
3.9051	535	0.2970	-
3.9416	540	0.2349	-
3.9781	545	0.1915	-
4.0	548	-	0.3205
4.0146	550	0.1950	-
4.0511	555	0.2137	-
4.0876	560	0.2170	-
4.1241	565	0.3080	-
4.1606	570	0.2038	-
4.1971	575	0.1622	-
4.2336	580	0.1701	-
4.2701	585	0.1806	-
4.3066	590	0.2083	-
4.3431	595	0.2606	-
4.3796	600	0.4207	-
4.4161	605	0.2372	-
4.4526	610	0.1963	-
4.4891	615	0.1232	-
4.5255	620	0.1927	-
4.5620	625	0.2543	-
4.5985	630	0.2017	-
4.6350	635	0.2134	-
4.6715	640	0.2981	-
4.7080	645	0.2670	-
4.7445	650	0.2822	-
4.7810	655	0.2664	-
4.8175	660	0.2651	-
4.8540	665	0.1945	-
4.8905	670	0.2803	-
4.9270	675	0.3401	-
4.9635	680	0.2055	-
5.0	685	0.2808	0.3240

The bold row denotes the saved checkpoint.

Training Time

Training: 1.9 hours

Framework Versions

Python: 3.12.13
Sentence Transformers: 5.4.1
Transformers: 5.0.0
PyTorch: 2.10.0+cu128
Accelerate: 1.13.0
Datasets: 4.0.0
Tokenizers: 0.22.2

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",
}

MultipleNegativesRankingLoss

bibtex

@misc{oord2019representationlearningcontrastivepredictive,
      title={Representation Learning with Contrastive Predictive Coding},
      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
      year={2019},
      eprint={1807.03748},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/1807.03748},
}

	anchor	positive
type	string	string
details	min: 7 tokens mean: 26.25 tokens max: 70 tokens	min: 9 tokens mean: 37.84 tokens max: 111 tokens