This is a sentence-transformers model finetuned from google/embeddinggemma-300m on the ola-owo/big-five-personality-traits dataset. It maps sentences & paragraphs to a 5-dimensional dense vector space and can be used for feature extraction.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: google/embeddinggemma-300m
Maximum Sequence Length: 2048 tokens
Output Dimensionality: 5 dimensions
Similarity Function: Cosine Similarity
Supported Modality: Text
Training Dataset:
- ola-owo/big-five-personality-traits
Languages: en, es

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': 'Gemma3TextModel'})
  (1): Pooling({'embedding_dimension': 768, 'pooling_mode': 'mean', 'include_prompt': True})
  (2): Dense({'in_features': 768, 'out_features': 3072, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity', 'module_input_name': 'sentence_embedding', 'module_output_name': 'sentence_embedding'})
  (3): Dense({'in_features': 3072, 'out_features': 768, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity', 'module_input_name': 'sentence_embedding', 'module_output_name': 'sentence_embedding'})
  (4): Normalize({})
  (5): Dense({'in_features': 768, 'out_features': 5, 'bias': True, 'activation_function': 'torch.nn.modules.linear.Identity', 'module_input_name': 'sentence_embedding', 'module_output_name': 'sentence_embedding'})
)

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("ola-owo/big5-sentence-transformer-lora")
# Run inference
queries = [
    'Se siente más viva en entornos sociales y dinámicos.',
]
documents = [
    'A menudo no cumple con los plazos y deja tareas incompletas, sin mostrar una gran responsabilidad.',
    'Muestra un comportamiento tranquilo y reservado, con momentos de entusiasmo social.',
    'Demonstrates flexible thinking that adapts between conventional and innovative approaches as needed.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 5] [3, 5]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[-0.0554, -0.9912, -0.4846]])

Training Details

Training Dataset

ola-owo/big-five-personality-traits

Dataset: ola-owo/big-five-personality-traits
Size: 2,250 training samples
Columns: sentence and label
Approximate statistics based on the first 100 samples:
sentence label
type string list
modality text
details
min: 8 tokens
mean: 19.07 tokens
max: 31 tokens
size: 5 elements

Samples:

sentence	label
`Intentos de mantener el orden, pero a menudo prioriza la comodidad sobre la estructura.`	`[-1.0, 0.15000000000000002, -1.0, -1.0, -1.0]`
`Tiende a ser reservada y menos asertiva, y rara vez busca emociones o llamar la atención.`	`[-1.0, -1.0, 0.15000000000000002, -1.0, -1.0]`
`Strikes a balance between socializing and alone time, feeling content with either.`	`[-1.0, -1.0, 0.5, -1.0, -1.0]`

Loss: [object Object].MultiLabelBCEWithLogitsLoss

Evaluation Dataset

ola-owo/big-five-personality-traits

Dataset: ola-owo/big-five-personality-traits
Size: 250 evaluation samples
Columns: sentence and label
Approximate statistics based on the first 100 samples:
sentence label
type string list
modality text
details
min: 9 tokens
mean: 19.09 tokens
max: 39 tokens
size: 5 elements

Samples:

sentence	label
`Maintains excellent organization and pays strong attention to detail in nearly everything.`	`[-1.0, 0.85, -1.0, -1.0, -1.0]`
`Enjoys occasional company but generally prefers quieter, low-key environments.`	`[-1.0, -1.0, 0.15000000000000002, -1.0, -1.0]`
`Will sacrifice personal interests and endure unfairness to avoid conflict or help someone in need.`	`[-1.0, -1.0, -1.0, 1.0, -1.0]`

Loss: [object Object].MultiLabelBCEWithLogitsLoss

Training Hyperparameters

Non-Default Hyperparameters

num_train_epochs: 20
learning_rate: 9.697682561279368e-05
warmup_steps: 0.1
auto_find_batch_size: True

All Hyperparameters

Click to expand

per_device_train_batch_size: 8
num_train_epochs: 20
max_steps: -1
learning_rate: 9.697682561279368e-05
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_steps: 0.1
optim: adamw_torch_fused
optim_args: None
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
optim_target_modules: None
gradient_accumulation_steps: 1
average_tokens_across_devices: True
max_grad_norm: 1.0
label_smoothing_factor: 0.0
bf16: False
fp16: False
bf16_full_eval: False
fp16_full_eval: False
tf32: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
use_liger_kernel: False
liger_kernel_config: None
use_cache: False
neftune_noise_alpha: None
torch_empty_cache_steps: None
auto_find_batch_size: True
log_on_each_node: True
logging_nan_inf_filter: True
include_num_input_tokens_seen: no
log_level: passive
log_level_replica: warning
disable_tqdm: False
project: huggingface
trackio_space_id: None
trackio_bucket_id: None
trackio_static_space_id: None
per_device_eval_batch_size: 8
prediction_loss_only: True
eval_on_start: False
eval_do_concat_batches: True
eval_use_gather_object: False
eval_accumulation_steps: None
include_for_metrics: []
batch_eval_metrics: False
save_only_model: False
save_on_each_node: False
enable_jit_checkpoint: False
push_to_hub: False
hub_private_repo: None
hub_model_id: None
hub_strategy: every_save
hub_always_push: False
hub_revision: None
load_best_model_at_end: False
ignore_data_skip: False
restore_callback_states_from_checkpoint: False
full_determinism: False
seed: 42
data_seed: None
use_cpu: 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
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_pin_memory: True
dataloader_persistent_workers: False
dataloader_prefetch_factor: None
remove_unused_columns: True
label_names: None
train_sampling_strategy: random
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
ddp_static_graph: None
ddp_backend: None
ddp_timeout: 1800
fsdp: None
fsdp_config: None
deepspeed: None
debug: []
skip_memory_metrics: True
do_predict: False
resume_from_checkpoint: None
warmup_ratio: None
local_rank: -1
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	Validation Loss
1.0	282	0.6071	0.5476
2.0	564	0.5313	0.6066
3.0	846	0.5313	0.5253
4.0	1128	0.5084	0.5030
5.0	1410	0.4808	0.5161
6.0	1692	0.4409	0.4823
7.0	1974	0.4221	0.5273
8.0	2256	0.4089	0.4880
9.0	2538	0.3902	0.4851
10.0	2820	0.3838	0.4351
11.0	3102	0.3713	0.4120
12.0	3384	0.3635	0.4158
13.0	3666	0.3609	0.4347
14.0	3948	0.3585	0.4289
15.0	4230	0.3564	0.4252
16.0	4512	0.3559	0.4183
17.0	4794	0.3554	0.4448
18.0	5076	0.3500	0.4263
19.0	5358	0.3464	0.4124
20.0	5640	0.3432	0.4121

Training Time

Training: 1.6 hours

Framework Versions

Python: 3.14.6
Sentence Transformers: 5.6.0
Transformers: 5.12.1
PyTorch: 2.12.1+cu132
Accelerate: 1.14.0
Datasets: 5.0.0
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",
}

	sentence	label
type	string	list
modality	text
details	min: 8 tokens mean: 19.07 tokens max: 31 tokens	size: 5 elements