This is a sentence-transformers model finetuned from sentence-transformers/all-mpnet-base-v2. 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: sentence-transformers/all-mpnet-base-v2
Maximum Sequence Length: 384 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': 384, 'do_lower_case': False, 'architecture': 'MPNetModel'})
  (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): 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("zoharzaig/emoji-prediction-model")
# Run inference
sentences = [
    'Inspired by the history behind Norfolk Island’s flag.',
    "The flag of Norfolk Island emoji represents the unique flag of Norfolk Island, which is an external territory of Australia. It is used to symbolize the island's culture and identity.",
    'The gear emoji is commonly used to represent machinery, equipment, tools, or mechanics. It can also symbolize maintenance, repair, or work involving gears and mechanical parts.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000,  0.7065, -0.0235],
#         [ 0.7065,  1.0000, -0.0110],
#         [-0.0235, -0.0110,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 139,891 training samples
Columns: sentence_0 and sentence_1
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1
type string string
details
min: 6 tokens
mean: 12.16 tokens
max: 29 tokens
min: 22 tokens
mean: 46.23 tokens
max: 89 tokens

Samples:

sentence_0	sentence_1
`Lunch is scheduled for eleven today`	`The eleven o’clock emoji is used to indicate the time of 11:00 on a clock. It can be used to show that it is late morning, or to signify that an event is happening at this specific time. It can also be used in a more figurative sense to represent the idea of being right on time for something.`
`Just finished reading an inspiring article on trans rights.`	`The transgender symbol emoji is often used to represent individuals who identify as transgender or non-binary`
`I'm curious about the history behind Lesotho’s flag.`	`The flag of Lesotho represents the country of Lesotho in southern Africa. It is a tricolor flag of horizontal stripes with a blue triangle on the left side. The colors symbolize different aspects of the country's history and culture.`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim" 4}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 5
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: 16
per_device_eval_batch_size: 16
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: 5
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: 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: None
hub_always_push: False
hub_revision: None
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
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.0572	500	1.2611
0.1144	1000	1.0953
0.1715	1500	0.9964
0.2287	2000	0.9722
0.2859	2500	0.9712
0.3431	3000	0.918
0.4003	3500	0.9296
0.4575	4000	0.9069
0.5146	4500	0.9062
0.5718	5000	0.8788
0.6290	5500	0.895
0.6862	6000	0.8601
0.7434	6500	0.8461
0.8005	7000	0.8379
0.8577	7500	0.8209
0.9149	8000	0.8015
0.9721	8500	0.8103
1.0293	9000	0.7828
1.0865	9500	0.7064
1.1436	10000	0.6881
1.2008	10500	0.7004
1.2580	11000	0.7121
1.3152	11500	0.7222
1.3724	12000	0.7183
1.4296	12500	0.7024
1.4867	13000	0.7114
1.5439	13500	0.7115
1.6011	14000	0.6858
1.6583	14500	0.6944
1.7155	15000	0.6867
1.7726	15500	0.6776
1.8298	16000	0.7172
1.8870	16500	0.7086
1.9442	17000	0.6882
2.0014	17500	0.6788
2.0586	18000	0.5488
2.1157	18500	0.5428
2.1729	19000	0.5628
2.2301	19500	0.5524
2.2873	20000	0.5695
2.3445	20500	0.5708
2.4016	21000	0.5703
2.4588	21500	0.5512
2.5160	22000	0.5646
2.5732	22500	0.5753
2.6304	23000	0.5739
2.6876	23500	0.554
2.7447	24000	0.5744
2.8019	24500	0.5236
2.8591	25000	0.5471
2.9163	25500	0.5576
2.9735	26000	0.5601
3.0306	26500	0.5004
3.0878	27000	0.4471
3.1450	27500	0.4588
3.2022	28000	0.4439
3.2594	28500	0.4283
3.3166	29000	0.4452
3.3737	29500	0.4446
3.4309	30000	0.4413
3.4881	30500	0.4377
3.5453	31000	0.4504
3.6025	31500	0.4312
3.6597	32000	0.4397
3.7168	32500	0.4376
3.7740	33000	0.4596
3.8312	33500	0.4501
3.8884	34000	0.4338
3.9456	34500	0.4609
4.0027	35000	0.4476
4.0599	35500	0.3652
4.1171	36000	0.3506
4.1743	36500	0.3481
4.2315	37000	0.3805
4.2887	37500	0.3574
4.3458	38000	0.3622
4.4030	38500	0.3686
4.4602	39000	0.3572
4.5174	39500	0.3791
4.5746	40000	0.3736
4.6317	40500	0.3514
4.6889	41000	0.3682
4.7461	41500	0.3625
4.8033	42000	0.3601
4.8605	42500	0.3703
4.9177	43000	0.3783
4.9748	43500	0.3583

Framework Versions

Python: 3.9.6
Sentence Transformers: 5.0.0
Transformers: 4.53.2
PyTorch: 2.7.1
Accelerate: 1.9.0
Datasets: 4.0.0
Tokenizers: 0.21.2

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{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}
}

	sentence_0	sentence_1
type	string	string
details	min: 6 tokens mean: 12.16 tokens max: 29 tokens	min: 22 tokens mean: 46.23 tokens max: 89 tokens