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 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': 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("aleynahukmet/all-MiniLM-L6-v2-8-layers")
# Run inference
sentences = [
    'A black dog is drinking next to a brown and white dog that is looking at an orange ball in the lake, whilst a horse and rider passes behind.',
    'There are two people running around a track in lane three and the one wearing a blue shirt with a green thing over the eyes is just barely ahead of the guy wearing an orange shirt and sunglasses.',
    'the guy is dead',
]
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

Datasets: sts-dev and sts-test
Evaluated with EmbeddingSimilarityEvaluator

Metric	sts-dev	sts-test
pearson_cosine	0.8649	0.8203
spearman_cosine	0.8649	0.819

Knowledge Distillation

Evaluated with MSEEvaluator

Metric	Value
negative_mse	-0.0245

Training Details

Training Dataset

Unnamed Dataset

Size: 9,014,210 training samples
Columns: sentence and label
Approximate statistics based on the first 1000 samples:
sentence label
type string list
details
min: 4 tokens
mean: 12.24 tokens
max: 52 tokens
size: 384 elements

Samples:

sentence	label
`A person on a horse jumps over a broken down airplane.`	`[-0.009216307662427425, 0.003964003175497055, 0.04029734805226326, 0.0030935262329876423, -0.03516044840216637, ...]`
`Children smiling and waving at camera`	`[-0.03215238079428673, 0.06086821109056473, 0.013251038268208504, -0.017755677923560143, 0.07927625626325607, ...]`
`A boy is jumping on skateboard in the middle of a red bridge.`	`[-0.020561737939715385, -0.03641558438539505, -0.039370208978652954, -0.0975518748164177, 0.005307587794959545, ...]`

Loss: MSELoss

Evaluation Dataset

Unnamed Dataset

Size: 10,000 evaluation samples
Columns: sentence and label
Approximate statistics based on the first 1000 samples:
sentence label
type string list
details
min: 5 tokens
mean: 13.23 tokens
max: 57 tokens
size: 384 elements

Samples:

sentence	label
`Two women are embracing while holding to go packages.`	`[-0.007923883385956287, -0.024198176339268684, 0.034445445984601974, 0.036053989082574844, -0.06740871071815491, ...]`
`Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.`	`[-0.08869566023349762, 0.02789478376507759, 0.060685668140649796, -0.02580258436501026, 0.008359752595424652, ...]`
`A man selling donuts to a customer during a world exhibition event held in the city of Angeles`	`[0.027255145832896233, 0.07622072845697403, 0.025504805147647858, -0.0542026124894619, -0.052822694182395935, ...]`

Loss: MSELoss

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 64
per_device_eval_batch_size: 64
learning_rate: 0.0001
num_train_epochs: 1
warmup_ratio: 0.1
fp16: True
load_best_model_at_end: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 64
per_device_eval_batch_size: 64
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: 0.0001
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: 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
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss	sts-dev_spearman_cosine	negative_mse	sts-test_spearman_cosine
0	0	-	-	0.7048	-0.3846	-
0.0071	1000	0.0032	-	-	-	-
0.0142	2000	0.0023	-	-	-	-
0.0213	3000	0.0019	-	-	-	-
0.0284	4000	0.0017	-	-	-	-
0.0355	5000	0.0015	0.0013	0.8149	-0.1309	-
0.0426	6000	0.0014	-	-	-	-
0.0497	7000	0.0012	-	-	-	-
0.0568	8000	0.0011	-	-	-	-
0.0639	9000	0.001	-	-	-	-
0.0710	10000	0.001	0.0008	0.8495	-0.0754	-
0.0781	11000	0.0009	-	-	-	-
0.0852	12000	0.0008	-	-	-	-
0.0923	13000	0.0008	-	-	-	-
0.0994	14000	0.0007	-	-	-	-
0.1065	15000	0.0007	0.0005	0.8569	-0.0528	-
0.1136	16000	0.0007	-	-	-	-
0.1207	17000	0.0007	-	-	-	-
0.1278	18000	0.0006	-	-	-	-
0.1349	19000	0.0006	-	-	-	-
0.1420	20000	0.0006	0.0004	0.8589	-0.0438	-
0.1491	21000	0.0006	-	-	-	-
0.1562	22000	0.0006	-	-	-	-
0.1633	23000	0.0006	-	-	-	-
0.1704	24000	0.0006	-	-	-	-
0.1775	25000	0.0005	0.0004	0.8608	-0.0392	-
0.1846	26000	0.0005	-	-	-	-
0.1917	27000	0.0005	-	-	-	-
0.1988	28000	0.0005	-	-	-	-
0.2059	29000	0.0005	-	-	-	-
0.2130	30000	0.0005	0.0004	0.8619	-0.0363	-
0.2201	31000	0.0005	-	-	-	-
0.2272	32000	0.0005	-	-	-	-
0.2343	33000	0.0005	-	-	-	-
0.2414	34000	0.0005	-	-	-	-
0.2485	35000	0.0005	0.0003	0.8619	-0.0343	-
0.2556	36000	0.0005	-	-	-	-
0.2627	37000	0.0005	-	-	-	-
0.2698	38000	0.0005	-	-	-	-
0.2769	39000	0.0005	-	-	-	-
0.2840	40000	0.0005	0.0003	0.8613	-0.0329	-
0.2911	41000	0.0005	-	-	-	-
0.2982	42000	0.0005	-	-	-	-
0.3053	43000	0.0005	-	-	-	-
0.3124	44000	0.0005	-	-	-	-
0.3195	45000	0.0005	0.0003	0.8633	-0.0316	-
0.3266	46000	0.0005	-	-	-	-
0.3337	47000	0.0005	-	-	-	-
0.3408	48000	0.0005	-	-	-	-
0.3479	49000	0.0004	-	-	-	-
0.3550	50000	0.0004	0.0003	0.8631	-0.0306	-
0.3621	51000	0.0004	-	-	-	-
0.3692	52000	0.0004	-	-	-	-
0.3763	53000	0.0004	-	-	-	-
0.3834	54000	0.0004	-	-	-	-
0.3905	55000	0.0004	0.0003	0.8635	-0.0297	-
0.3976	56000	0.0004	-	-	-	-
0.4047	57000	0.0004	-	-	-	-
0.4118	58000	0.0004	-	-	-	-
0.4189	59000	0.0004	-	-	-	-
0.4260	60000	0.0004	0.0003	0.8640	-0.0290	-
0.4331	61000	0.0004	-	-	-	-
0.4402	62000	0.0004	-	-	-	-
0.4473	63000	0.0004	-	-	-	-
0.4544	64000	0.0004	-	-	-	-
0.4615	65000	0.0004	0.0003	0.8644	-0.0285	-
0.4686	66000	0.0004	-	-	-	-
0.4757	67000	0.0004	-	-	-	-
0.4828	68000	0.0004	-	-	-	-
0.4899	69000	0.0004	-	-	-	-
0.4970	70000	0.0004	0.0003	0.8641	-0.0280	-
0.5041	71000	0.0004	-	-	-	-
0.5112	72000	0.0004	-	-	-	-
0.5183	73000	0.0004	-	-	-	-
0.5254	74000	0.0004	-	-	-	-
0.5325	75000	0.0004	0.0003	0.8648	-0.0276	-
0.5396	76000	0.0004	-	-	-	-
0.5467	77000	0.0004	-	-	-	-
0.5538	78000	0.0004	-	-	-	-
0.5609	79000	0.0004	-	-	-	-
0.5680	80000	0.0004	0.0003	0.8644	-0.0271	-
0.5751	81000	0.0004	-	-	-	-
0.5822	82000	0.0004	-	-	-	-
0.5893	83000	0.0004	-	-	-	-
0.5964	84000	0.0004	-	-	-	-
0.6035	85000	0.0004	0.0003	0.8648	-0.0267	-
0.6106	86000	0.0004	-	-	-	-
0.6177	87000	0.0004	-	-	-	-
0.6248	88000	0.0004	-	-	-	-
0.6319	89000	0.0004	-	-	-	-
0.6390	90000	0.0004	0.0003	0.8645	-0.0264	-
0.6461	91000	0.0004	-	-	-	-
0.6532	92000	0.0004	-	-	-	-
0.6603	93000	0.0004	-	-	-	-
0.6674	94000	0.0004	-	-	-	-
0.6745	95000	0.0004	0.0003	0.8643	-0.0261	-
0.6816	96000	0.0004	-	-	-	-
0.6887	97000	0.0004	-	-	-	-
0.6958	98000	0.0004	-	-	-	-
0.7029	99000	0.0004	-	-	-	-
0.7100	100000	0.0004	0.0003	0.8643	-0.0259	-
0.7171	101000	0.0004	-	-	-	-
0.7242	102000	0.0004	-	-	-	-
0.7313	103000	0.0004	-	-	-	-
0.7384	104000	0.0004	-	-	-	-
0.7455	105000	0.0004	0.0003	0.8646	-0.0257	-
0.7526	106000	0.0004	-	-	-	-
0.7597	107000	0.0004	-	-	-	-
0.7668	108000	0.0004	-	-	-	-
0.7739	109000	0.0004	-	-	-	-
0.7810	110000	0.0004	0.0003	0.8637	-0.0254	-
0.7881	111000	0.0004	-	-	-	-
0.7952	112000	0.0004	-	-	-	-
0.8023	113000	0.0004	-	-	-	-
0.8094	114000	0.0004	-	-	-	-
0.8165	115000	0.0004	0.0003	0.8643	-0.0252	-
0.8236	116000	0.0004	-	-	-	-
0.8307	117000	0.0004	-	-	-	-
0.8378	118000	0.0004	-	-	-	-
0.8449	119000	0.0004	-	-	-	-
0.8520	120000	0.0004	0.0003	0.8645	-0.0250	-
0.8591	121000	0.0004	-	-	-	-
0.8662	122000	0.0004	-	-	-	-
0.8733	123000	0.0004	-	-	-	-
0.8804	124000	0.0004	-	-	-	-
0.8875	125000	0.0004	0.0002	0.8646	-0.0248	-
0.8946	126000	0.0004	-	-	-	-
0.9017	127000	0.0004	-	-	-	-
0.9088	128000	0.0004	-	-	-	-
0.9159	129000	0.0004	-	-	-	-
0.9230	130000	0.0004	0.0002	0.8647	-0.0247	-
0.9301	131000	0.0004	-	-	-	-
0.9372	132000	0.0004	-	-	-	-
0.9443	133000	0.0004	-	-	-	-
0.9514	134000	0.0004	-	-	-	-
0.9585	135000	0.0004	0.0002	0.8646	-0.0246	-
0.9656	136000	0.0004	-	-	-	-
0.9727	137000	0.0004	-	-	-	-
0.9798	138000	0.0004	-	-	-	-
0.9869	139000	0.0004	-	-	-	-
0.994	140000	0.0004	0.0002	0.8649	-0.0245	-
1.0	140848	-	-	-	-	0.8190

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.12.4
Sentence Transformers: 3.3.1
Transformers: 4.44.2
PyTorch: 2.4.1+cu121
Accelerate: 1.0.1
Datasets: 2.19.0
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",
}

MSELoss

bibtex

@inproceedings{reimers-2020-multilingual-sentence-bert,
    title = "Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2020",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/2004.09813",
}

	sentence	label
type	string	list
details	min: 4 tokens mean: 12.24 tokens max: 52 tokens	size: 384 elements