Lampistero

This is a sentence-transformers model finetuned from jinaai/jina-embeddings-v3 on the json dataset. It maps sentences & paragraphs to a 1024-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: jinaai/jina-embeddings-v3
Maximum Sequence Length: 8194 tokens
Output Dimensionality: 1024 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- json
Language: es
License: apache-2.0

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (transformer): Transformer(
    (auto_model): XLMRobertaLoRA(
      (roberta): XLMRobertaModel(
        (embeddings): XLMRobertaEmbeddings(
          (word_embeddings): ParametrizedEmbedding(
            250002, 1024, padding_idx=1
            (parametrizations): ModuleDict(
              (weight): ParametrizationList(
                (0): LoRAParametrization()
              )
            )
          )
          (token_type_embeddings): ParametrizedEmbedding(
            1, 1024
            (parametrizations): ModuleDict(
              (weight): ParametrizationList(
                (0): LoRAParametrization()
              )
            )
          )
        )
        (emb_drop): Dropout(p=0.1, inplace=False)
        (emb_ln): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
        (encoder): XLMRobertaEncoder(
          (layers): ModuleList(
            (0-23): 24 x Block(
              (mixer): MHA(
                (rotary_emb): RotaryEmbedding()
                (Wqkv): ParametrizedLinearResidual(
                  in_features=1024, out_features=3072, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
                (inner_attn): FlashSelfAttention(
                  (drop): Dropout(p=0.1, inplace=False)
                )
                (inner_cross_attn): FlashCrossAttention(
                  (drop): Dropout(p=0.1, inplace=False)
                )
                (out_proj): ParametrizedLinear(
                  in_features=1024, out_features=1024, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
              )
              (dropout1): Dropout(p=0.1, inplace=False)
              (drop_path1): StochasticDepth(p=0.0, mode=row)
              (norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
              (mlp): Mlp(
                (fc1): ParametrizedLinear(
                  in_features=1024, out_features=4096, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
                (fc2): ParametrizedLinear(
                  in_features=4096, out_features=1024, bias=True
                  (parametrizations): ModuleDict(
                    (weight): ParametrizationList(
                      (0): LoRAParametrization()
                    )
                  )
                )
              )
              (dropout2): Dropout(p=0.1, inplace=False)
              (drop_path2): StochasticDepth(p=0.0, mode=row)
              (norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
            )
          )
        )
        (pooler): XLMRobertaPooler(
          (dense): ParametrizedLinear(
            in_features=1024, out_features=1024, bias=True
            (parametrizations): ModuleDict(
              (weight): ParametrizationList(
                (0): LoRAParametrization()
              )
            )
          )
          (activation): Tanh()
        )
      )
    )
  )
  (pooler): Pooling({'word_embedding_dimension': 1024, '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})
  (normalizer): 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.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("csanz91/lampistero_rag_embeddings")
# Run inference
sentences = [
    '¿Qué porcentaje de aumento salarial reclamaba el Sindicato Minero en el conflicto de Utrillas que llevó a plantear la huelga del 12 de octubre de 1930?',
    'El Sindicato Minero reclamaba un aumento del 20% los sueldos en el conflicto de Utrillas.',
    'Antonio Gargallo.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Information Retrieval

Dataset: dim_1024
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 1024
}
```

Metric	Value
cosine_accuracy@1	0.7803
cosine_accuracy@3	0.8884
cosine_accuracy@5	0.904
cosine_accuracy@10	0.9234
cosine_precision@1	0.7803
cosine_precision@3	0.2961
cosine_precision@5	0.1808
cosine_precision@10	0.0923
cosine_recall@1	0.7803
cosine_recall@3	0.8884
cosine_recall@5	0.904
cosine_recall@10	0.9234
cosine_ndcg@10	0.8576
cosine_mrr@10	0.8359
cosine_map@100	0.8374

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 768
}
```

Metric	Value
cosine_accuracy@1	0.7827
cosine_accuracy@3	0.8877
cosine_accuracy@5	0.9034
cosine_accuracy@10	0.9246
cosine_precision@1	0.7827
cosine_precision@3	0.2959
cosine_precision@5	0.1807
cosine_precision@10	0.0925
cosine_recall@1	0.7827
cosine_recall@3	0.8877
cosine_recall@5	0.9034
cosine_recall@10	0.9246
cosine_ndcg@10	0.8588
cosine_mrr@10	0.8372
cosine_map@100	0.8385

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 512
}
```

Metric	Value
cosine_accuracy@1	0.7797
cosine_accuracy@3	0.8859
cosine_accuracy@5	0.901
cosine_accuracy@10	0.9228
cosine_precision@1	0.7797
cosine_precision@3	0.2953
cosine_precision@5	0.1802
cosine_precision@10	0.0923
cosine_recall@1	0.7797
cosine_recall@3	0.8859
cosine_recall@5	0.901
cosine_recall@10	0.9228
cosine_ndcg@10	0.8564
cosine_mrr@10	0.8347
cosine_map@100	0.8362

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 256
}
```

Metric	Value
cosine_accuracy@1	0.7707
cosine_accuracy@3	0.8823
cosine_accuracy@5	0.9016
cosine_accuracy@10	0.9191
cosine_precision@1	0.7707
cosine_precision@3	0.2941
cosine_precision@5	0.1803
cosine_precision@10	0.0919
cosine_recall@1	0.7707
cosine_recall@3	0.8823
cosine_recall@5	0.9016
cosine_recall@10	0.9191
cosine_ndcg@10	0.8512
cosine_mrr@10	0.8287
cosine_map@100	0.8303

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 128
}
```

Metric	Value
cosine_accuracy@1	0.7604
cosine_accuracy@3	0.869
cosine_accuracy@5	0.8902
cosine_accuracy@10	0.9131
cosine_precision@1	0.7604
cosine_precision@3	0.2897
cosine_precision@5	0.178
cosine_precision@10	0.0913
cosine_recall@1	0.7604
cosine_recall@3	0.869
cosine_recall@5	0.8902
cosine_recall@10	0.9131
cosine_ndcg@10	0.8415
cosine_mrr@10	0.8181
cosine_map@100	0.82

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 64
}
```

Metric	Value
cosine_accuracy@1	0.7248
cosine_accuracy@3	0.8521
cosine_accuracy@5	0.8751
cosine_accuracy@10	0.8974
cosine_precision@1	0.7248
cosine_precision@3	0.284
cosine_precision@5	0.175
cosine_precision@10	0.0897
cosine_recall@1	0.7248
cosine_recall@3	0.8521
cosine_recall@5	0.8751
cosine_recall@10	0.8974
cosine_ndcg@10	0.8182
cosine_mrr@10	0.792
cosine_map@100	0.7938

Training Details

Training Dataset

json

Dataset: json
Size: 14,907 training samples
Columns: query and answer
Approximate statistics based on the first 1000 samples:
query answer
type string string
details
min: 9 tokens
mean: 25.88 tokens
max: 63 tokens

min: 3 tokens
mean: 34.09 tokens
max: 340 tokens

	query	answer
type	string	string
details	min: 9 tokens mean: 25.88 tokens max: 63 tokens	min: 3 tokens mean: 34.09 tokens max: 340 tokens

Samples:

query	answer
`En Valdeconejos, ¿cuál era la sociedad de agricultores en 1952?`	`En Valdeconejos, la sociedad de agricultores en 1952 era el Pósito de Agricultores.`
`¿Qué nombres de capataces se registran en el pueblo de Escucha en el año 1952?`	`En Escucha, en 1952, los capataces registrados son Peralta (Manuel) y Rodriguez (Gonzalo).`
`En el contexto de la minería, ¿qué implica 'despajar'?`	`'Despajar' se refiere a cribar a mano material y desechos para obtener las partes de carbón que hay en ellos.`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        1024,
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 64
per_device_eval_batch_size: 16
gradient_accumulation_steps: 32
learning_rate: 2e-05
num_train_epochs: 12
lr_scheduler_type: cosine
warmup_ratio: 0.1
tf32: True
load_best_model_at_end: True
optim: adamw_torch_fused
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 64
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 32
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 2e-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: 12
max_steps: -1
lr_scheduler_type: cosine
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: True
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}
tp_size: 0
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_fused
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
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
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	dim_1024_cosine_ndcg@10	dim_768_cosine_ndcg@10	dim_512_cosine_ndcg@10	dim_256_cosine_ndcg@10	dim_128_cosine_ndcg@10	dim_64_cosine_ndcg@10
1.0	8	-	0.7663	0.7676	0.7656	0.7626	0.7393	0.6969
1.2747	10	127.0406	-	-	-	-	-	-
2.0	16	-	0.8244	0.8240	0.8226	0.8172	0.8060	0.7775
2.5494	20	38.8995	-	-	-	-	-	-
3.0	24	-	0.8425	0.8426	0.8444	0.8373	0.8252	0.7996
3.8240	30	20.1528	-	-	-	-	-	-
4.0	32	-	0.8526	0.8520	0.8498	0.8456	0.8289	0.8037
5.0	40	14.0513	0.8550	0.8543	0.8517	0.8490	0.8368	0.8139
6.0	48	-	0.8572	0.8565	0.8557	0.8520	0.8404	0.8170
6.2747	50	13.364	-	-	-	-	-	-
7.0	56	-	0.8579	0.8576	0.8553	0.8514	0.8422	0.8180
7.5494	60	12.7986	-	-	-	-	-	-
8.0	64	-	0.8573	0.8580	0.8560	0.8523	0.8414	0.8178
8.8240	70	12.0091	-	-	-	-	-	-
9.0	72	-	0.8578	0.8586	0.8562	0.8519	0.8423	0.8184
10.0	80	10.9468	0.8583	0.8589	0.8565	0.8530	0.8413	0.8191
10.5494	84	-	0.8576	0.8588	0.8564	0.8512	0.8415	0.8182

Framework Versions

Python: 3.12.10
Sentence Transformers: 4.1.0
Transformers: 4.51.3
PyTorch: 2.7.0+cu126
Accelerate: 1.7.0
Datasets: 3.6.0
Tokenizers: 0.21.1

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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@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}
}

csanz91
/

lampistero_rag_embeddings

Lampistero

Model Details

Model Description

Model Sources

Full Model Architecture

Usage

Direct Usage (Sentence Transformers)

Evaluation

Metrics

Information Retrieval

Information Retrieval

Information Retrieval

Information Retrieval

Information Retrieval

Information Retrieval

Training Details

Training Dataset

json

Training Hyperparameters

Non-Default Hyperparameters

All Hyperparameters

Training Logs

Framework Versions

Citation

BibTeX

Sentence Transformers

MatryoshkaLoss

MultipleNegativesRankingLoss

Model tree for csanz91/lampistero_rag_embeddings

Evaluation results