ktcapraz/embeddinggemma-300m-german-search-categories

embeddinggemma-300m fine-tuned on German Search Categories

This is a sentence-transformers model finetuned from google/embeddinggemma-300m on the search-categories-german-triplets dataset. 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: google/embeddinggemma-300m <!-- at revision c5cfa06e5e282a820e85d57f7fb053207494f41d -->
• —Maximum Sequence Length: 2048 tokens
• —Output Dimensionality: 768 dimensions
• —Similarity Function: Cosine Similarity
• —Training Dataset:
• —search-categories-german-triplets
• —Language: de
• —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(
  (0): Transformer({'max_seq_length': 2048, 'do_lower_case': False, 'architecture': 'Gemma3TextModel'})
  (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): Dense({'in_features': 768, 'out_features': 3072, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
  (3): Dense({'in_features': 3072, 'out_features': 768, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
  (4): 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("ktcapraz/embeddinggemma-300m-german-search-categories")
# Run inference
queries = [
    "Wann endet die Verhandlungsfrist f\u00fcr Spielerlisten 2024?",
]
documents = [
    'A broad, general question, related to a specific time.',
    'A specific question about a person or entity, with no time constraint.',
    'A keyword-based search for a broad topic, with no time constraint.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 768] [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.7721, 0.3525, 0.1172]])

<!--

Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details> -->

<!--

Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details> -->

<!--

Out-of-Scope Use

List how the model may foreseeably be misused and address what users ought not to do with the model. -->

Evaluation

Metrics

Information Retrieval

• —Datasets: german-search-cats-eval and german-search-cats-test
• —Evaluated with <code>InformationRetrievalEvaluator</code>

Information Retrieval

• —Dataset: german-search-cats-eval
• —Evaluated with <code>InformationRetrievalEvaluator</code>

<!--

Bias, Risks and Limitations

What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model. -->

<!--

Recommendations

What are recommendations with respect to the foreseeable issues? For example, filtering explicit content. -->

Training Details

Training Dataset

search-categories-german-triplets

• —Dataset: search-categories-german-triplets at bbf15f6
• —Size: 73,141 training samples
• —Columns: <code>anchor</code>, <code>positive</code>, and <code>negatives</code>
• —Approximate statistics based on the first 1000 samples: | | anchor | positive | negatives | |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------| | type | string | string | list | | details | <ul><li>min: 4 tokens</li><li>mean: 12.97 tokens</li><li>max: 25 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 16.08 tokens</li><li>max: 18 tokens</li></ul> | <ul><li>size: 3 elements</li></ul> |
• —Samples: | anchor | positive | negatives | |:------------------------------------------------------------------|:----------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | <code>Fleischerzeugung Umweltkosten Deutschland</code> | <code>A keyword-based search for a broad topic, with no time constraint.</code> | <code>['A keyword-based search for a broad topic, related to a specific time.', 'A broad, general question, with no time constraint.', 'A specific keyword search for a person or entity, with no time constraint.']</code> | | <code>Schule Rösrath Mpox Schließung</code> | <code>A specific keyword search for a person or entity, with no time constraint.</code> | <code>['A specific keyword search for a person or entity, related to a specific time.', 'A specific question about a person or entity, with no time constraint.', 'A keyword-based search for a broad topic, with no time constraint.']</code> | | <code>Was ändert sich bei der Passbeantragung ab Mai 2024?</code> | <code>A broad, general question, related to a specific time.</code> | <code>['A keyword-based search for a broad topic, related to a specific time.', 'A broad, general question, with no time constraint.', 'A specific question about a person or entity, related to a specific time.']</code> |
• —Loss: <code>CachedMultipleNegativesRankingLoss</code> with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "mini_batch_size": 8,
      "gather_across_devices": false
  }

Evaluation Dataset

search-categories-german-triplets

• —Dataset: search-categories-german-triplets at bbf15f6
• —Size: 15,674 evaluation samples
• —Columns: <code>anchor</code>, <code>positive</code>, and <code>negatives</code>
• —Approximate statistics based on the first 1000 samples: | | anchor | positive | negatives | |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------| | type | string | string | list | | details | <ul><li>min: 4 tokens</li><li>mean: 12.59 tokens</li><li>max: 26 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 16.21 tokens</li><li>max: 18 tokens</li></ul> | <ul><li>size: 3 elements</li></ul> |
• —Samples: | anchor | positive | negatives | |:--------------------------------------------------------------------|:-------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | <code>Weihnachtsgeschäft Einzelhandel November Dezember 2024</code> | <code>A specific keyword search for a person or entity, related to a specific time.</code> | <code>['A keyword-based search for a broad topic, related to a specific time.', 'A specific keyword search for a person or entity, with no time constraint.', 'A specific question about a person or entity, related to a specific time.']</code> | | <code>Wie reagiert die Union auf Mützenichs Vorstoß?</code> | <code>A specific question about a person or entity, with no time constraint.</code> | <code>['A specific keyword search for a person or entity, with no time constraint.', 'A specific question about a person or entity, related to a specific time.', 'A broad, general question, with no time constraint.']</code> | | <code>Al-Manar TV Sperrverfügung Telekom</code> | <code>A specific keyword search for a person or entity, with no time constraint.</code> | <code>['A specific keyword search for a person or entity, related to a specific time.', 'A specific question about a person or entity, with no time constraint.', 'A keyword-based search for a broad topic, with no time constraint.']</code> |
• —Loss: <code>CachedMultipleNegativesRankingLoss</code> with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "mini_batch_size": 8,
      "gather_across_devices": false
  }

Training Hyperparameters

Non-Default Hyperparameters

• —eval_strategy: steps
• —per_device_train_batch_size: 128
• —per_device_eval_batch_size: 128
• —learning_rate: 1e-05
• —num_train_epochs: 1
• —warmup_ratio: 0.1
• —fp16: True
• —prompts: {'anchor': 'task: search result | query: ', 'positive': 'task: classification | query: '}
• —batch_sampler: no_duplicates

All Hyperparameters

<details><summary>Click to expand</summary>

• —overwrite_output_dir: False
• —do_predict: False
• —eval_strategy: steps
• —prediction_loss_only: True
• —per_device_train_batch_size: 128
• —per_device_eval_batch_size: 128
• —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: 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: 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: False
• —ignore_data_skip: False
• —fsdp: []
• —fsdp_min_num_params: 0
• —fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
• —fsdp_transformer_layer_cls_to_wrap: None
• —accelerator_config: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
• —parallelism_config: None
• —deepspeed: None
• —label_smoothing_factor: 0.0
• —optim: adamwtorchfused
• —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: {'anchor': 'task: search result | query: ', 'positive': 'task: classification | query: '}
• —batch_sampler: no_duplicates
• —multi_dataset_batch_sampler: proportional
• —router_mapping: {}
• —learning_rate_mapping: {}

</details>

Training Logs

Framework Versions

• —Python: 3.10.18
• —Sentence Transformers: 5.1.0
• —Transformers: 4.56.1
• —PyTorch: 2.8.0+cu128
• —Accelerate: 1.10.1
• —Datasets: 4.1.0
• —Tokenizers: 0.22.0

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",
}

CachedMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

<!--

Glossary

Clearly define terms in order to be accessible across audiences. -->

<!--

Model Card Authors

Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction. -->

<!--

Model Card Contact

Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors. -->