finetune_21_08_2025_18_35_25 / README.md

Add new SentenceTransformer model

9125365 verified 4 months ago

25.8 kB

	---
	tags:
	- sentence-transformers
	- sentence-similarity
	- feature-extraction
	- dense
	- generated_from_trainer
	- dataset_size:19964
	- loss:MultipleNegativesRankingLoss
	base_model: Snowflake/snowflake-arctic-embed-m-v2.0
	widget:
	- source_sentence: 'Kollegin hat Probleme mit dem Login zu '
	sentences:
	- Alle genannten Kinder gab es in kitaplus. Bei einem musste nur eine neue BI angelegt
	werden, bei den anderen muss der Vertrag in einer anderen Kita rückgängig gemacht
	werden, damit es in kitaplus in dieser Einrichtung aus der Liste der Absagen genommen
	werden kann.
	- Der Bereich ist aktuell noch nicht sichtbar.
	- muss mit dem Rentamt geklärt werden
	- source_sentence: Benutzer möchte einen Kollegen nur für die Dokumentenbibliothek
	anlegen.
	sentences:
	- Rücksprache mit Entwickler.
	- Sie muss den Regler auf Anzahl stellen
	- Zusammen die Rolle gewählt und dort dann in den individuellen Rechten alles auf
	lesend bzw. ausblenden gestellt, außer die Bibliothek.
	- source_sentence: Ist es richtig so, dass Mitarbeiter, wenn sie nach einer gewissen
	Zeit wieder in die Einrichtung kommen, erneut angelegt werden müssen?
	sentences:
	- Userin an den Träger verwiesen, dieser kann bei ihr ein neues Passwort setzen.
	- Ja, das ist korrekt so.
	- Userin muss erst rechts über das 3-Punkte-menü die "Anmeldedaten zusammenführen".
	Danach muss man in den angelegten BI die Gruppenform des Anmeldeportals angeben.
	- source_sentence: Userin kann die Öffnungszeiten der Einrichtung nicht bearbeiten.
	sentences:
	- informiert, dass es keinen Testzugang gibt, aber Handbücher und Hilfen in zur
	Verfügung stehen, wenn die Schnittstelle eingerichtet wurde.
	- Bereits bekannt, die Kollegen sind schon dabei den Fehler zu beheben.
	- Userin darf dies mit der Rolle nicht.
	- source_sentence: fragt wie der Stand zu dem aktuellen Problem ist
	sentences:
	- Userin muss sich an die Bistums IT wenden.
	- In Klärung mit der Kollegin - Das Problem liegt leider an deren Betreiber. Die
	sind aber informiert und arbeiten bereits daran
	- findet diese in der Übersicht der Gruppen.
	pipeline_tag: sentence-similarity
	library_name: sentence-transformers
	metrics:
	- cosine_accuracy@1
	- cosine_accuracy@3
	- cosine_accuracy@5
	- cosine_accuracy@10
	- cosine_precision@1
	- cosine_precision@3
	- cosine_precision@5
	- cosine_precision@10
	- cosine_recall@1
	- cosine_recall@3
	- cosine_recall@5
	- cosine_recall@10
	- cosine_ndcg@10
	- cosine_mrr@10
	- cosine_map@100
	model-index:
	- name: SentenceTransformer based on Snowflake/snowflake-arctic-embed-m-v2.0
	results:
	- task:
	type: information-retrieval
	name: Information Retrieval
	dataset:
	name: Snowflake/snowflake arctic embed m v2.0
	type: Snowflake/snowflake-arctic-embed-m-v2.0
	metrics:
	- type: cosine_accuracy@1
	value: 0.19708029197080293
	name: Cosine Accuracy@1
	- type: cosine_accuracy@3
	value: 0.7226277372262774
	name: Cosine Accuracy@3
	- type: cosine_accuracy@5
	value: 0.8029197080291971
	name: Cosine Accuracy@5
	- type: cosine_accuracy@10
	value: 0.8759124087591241
	name: Cosine Accuracy@10
	- type: cosine_precision@1
	value: 0.19708029197080293
	name: Cosine Precision@1
	- type: cosine_precision@3
	value: 0.44525547445255476
	name: Cosine Precision@3
	- type: cosine_precision@5
	value: 0.46277372262773725
	name: Cosine Precision@5
	- type: cosine_precision@10
	value: 0.43576642335766425
	name: Cosine Precision@10
	- type: cosine_recall@1
	value: 0.008762531776700945
	name: Cosine Recall@1
	- type: cosine_recall@3
	value: 0.09805489105617915
	name: Cosine Recall@3
	- type: cosine_recall@5
	value: 0.1603290464604333
	name: Cosine Recall@5
	- type: cosine_recall@10
	value: 0.23250747987759582
	name: Cosine Recall@10
	- type: cosine_ndcg@10
	value: 0.4532269034566889
	name: Cosine Ndcg@10
	- type: cosine_mrr@10
	value: 0.47734040088054697
	name: Cosine Mrr@10
	- type: cosine_map@100
	value: 0.2936078777768552
	name: Cosine Map@100
	---

	# SentenceTransformer based on Snowflake/snowflake-arctic-embed-m-v2.0

	This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Snowflake/snowflake-arctic-embed-m-v2.0](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v2.0) on the train 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: [Snowflake/snowflake-arctic-embed-m-v2.0](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v2.0) <!-- at revision 95c2741480856aa9666782eb4afe11959938017f -->
	- Maximum Sequence Length: 8192 tokens
	- Output Dimensionality: 768 dimensions
	- Similarity Function: Cosine Similarity
	- Training Dataset:
	- train
	<!-- - Language: Unknown -->
	<!-- - License: Unknown -->

	### Model Sources

	- Documentation: [Sentence Transformers Documentation](https://sbert.net)
	- Repository: [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
	- Hugging Face: [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)

	### Full Model Architecture

	```
	SentenceTransformer(
	(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False, 'architecture': 'GteModel'})
	(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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:

	```bash
	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("BjarneNPO/finetune_21_08_2025_18_35_25")
	# Run inference
	queries = [
	"fragt wie der Stand zu dem aktuellen Problem ist",
	]
	documents = [
	'In Klärung mit der Kollegin - Das Problem liegt leider an deren Betreiber. Die sind aber informiert und arbeiten bereits daran',
	'findet diese in der Übersicht der Gruppen.',
	'Userin muss sich an die Bistums IT wenden.',
	]
	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([[0.2744, 0.0387, 0.0701]])
	```

	<!--
	### 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

	* Dataset: `Snowflake/snowflake-arctic-embed-m-v2.0`
	* Evaluated with <code>scripts.InformationRetrievalEvaluatorCustom.InformationRetrievalEvaluatorCustom</code> with these parameters:
	```json
	{
	"query_prompt_name": "query",
	"corpus_prompt_name": "query"
	}
	```

	\| Metric \| Value \|
	\|:--------------------\|:-----------\|
	\| cosine_accuracy@1 \| 0.1971 \|
	\| cosine_accuracy@3 \| 0.7226 \|
	\| cosine_accuracy@5 \| 0.8029 \|
	\| cosine_accuracy@10 \| 0.8759 \|
	\| cosine_precision@1 \| 0.1971 \|
	\| cosine_precision@3 \| 0.4453 \|
	\| cosine_precision@5 \| 0.4628 \|
	\| cosine_precision@10 \| 0.4358 \|
	\| cosine_recall@1 \| 0.0088 \|
	\| cosine_recall@3 \| 0.0981 \|
	\| cosine_recall@5 \| 0.1603 \|
	\| cosine_recall@10 \| 0.2325 \|
	\| cosine_ndcg@10 \| 0.4532 \|
	\| cosine_mrr@10 \| 0.4773 \|
	\| cosine_map@100 \| 0.2936 \|

	<!--
	## 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

	#### train

	* Dataset: train
	* Size: 19,964 training samples
	* Columns: <code>query</code> and <code>answer</code>
	* Approximate statistics based on the first 1000 samples:
	\| \| query \| answer \|
	\|:--------\|:-----------------------------------------------------------------------------------\|:-----------------------------------------------------------------------------------\|
	\| type \| string \| string \|
	\| details \| <ul><li>min: 4 tokens</li><li>mean: 27.77 tokens</li><li>max: 615 tokens</li></ul> \| <ul><li>min: 3 tokens</li><li>mean: 22.87 tokens</li><li>max: 151 tokens</li></ul> \|
	* Samples:
	\| query \| answer \|
	\|:------------------------------------------------------------------------------------------------------------------------------------------------------------\|:--------------------------------------------------------------------------------------------\|
	\| <code>Wie kann man die Jahresurlaubsübersicht exportieren?</code> \| <code>über das 3 Punkte Menü rechts oben. Mitarbeiter auswählen und exportieren</code> \|
	\| <code>1. Vertragsabschlüsse werden nicht übertragen <br>2. Kinder kommen nicht von nach <br>3. Absage kann bei Portalstatus nicht erstellt werden.</code> \| <code>Ticket <br>Userin gebeten sich an den Support zu wenden, da der Fehler liegt.</code> \|
	\| <code>Wird im Anmeldeportal nicht gefunden.</code> \| <code>Die Schnittstelle war noch nicht aktiviert und Profil ebenfalls nicht.</code> \|
	* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
	```json
	{
	"scale": 20.0,
	"similarity_fct": "cos_sim",
	"gather_across_devices": false
	}
	```

	### Training Hyperparameters
	#### Non-Default Hyperparameters

	- `eval_strategy`: epoch
	- `per_device_train_batch_size`: 64
	- `per_device_eval_batch_size`: 64
	- `gradient_accumulation_steps`: 4
	- `learning_rate`: 2e-05
	- `num_train_epochs`: 10
	- `lr_scheduler_type`: cosine
	- `warmup_ratio`: 0.1
	- `bf16`: True
	- `tf32`: True
	- `load_best_model_at_end`: True
	- `batch_sampler`: no_duplicates

	#### All Hyperparameters
	<details><summary>Click to expand</summary>

	- `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`: 64
	- `per_gpu_train_batch_size`: None
	- `per_gpu_eval_batch_size`: None
	- `gradient_accumulation_steps`: 4
	- `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`: 10
	- `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`: True
	- `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}
	- `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
	- `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`: no_duplicates
	- `multi_dataset_batch_sampler`: proportional
	- `router_mapping`: {}
	- `learning_rate_mapping`: {}

	</details>

	### Training Logs
	\| Epoch \| Step \| Training Loss \| Snowflake/snowflake-arctic-embed-m-v2.0_cosine_ndcg@10 \|
	\|:-------:\|:-------:\|:-------------:\|:------------------------------------------------------:\|
	\| 0.1282 \| 10 \| 3.4817 \| - \|
	\| 0.2564 \| 20 \| 3.3293 \| - \|
	\| 0.3846 \| 30 \| 3.2454 \| - \|
	\| 0.5128 \| 40 \| 2.9853 \| - \|
	\| 0.6410 \| 50 \| 2.8363 \| - \|
	\| 0.7692 \| 60 \| 2.6833 \| - \|
	\| 0.8974 \| 70 \| 2.5117 \| - \|
	\| 1.0 \| 78 \| - \| 0.5070 \|
	\| 1.0256 \| 80 \| 2.297 \| - \|
	\| 1.1538 \| 90 \| 2.2586 \| - \|
	\| 1.2821 \| 100 \| 2.1379 \| - \|
	\| 1.4103 \| 110 \| 2.1199 \| - \|
	\| 1.5385 \| 120 \| 2.0054 \| - \|
	\| 1.6667 \| 130 \| 1.9546 \| - \|
	\| 1.7949 \| 140 \| 1.8525 \| - \|
	\| 1.9231 \| 150 \| 1.8471 \| - \|
	\| 2.0 \| 156 \| - \| 0.4817 \|
	\| 2.0513 \| 160 \| 1.6686 \| - \|
	\| 2.1795 \| 170 \| 1.7224 \| - \|
	\| 2.3077 \| 180 \| 1.7122 \| - \|
	\| 2.4359 \| 190 \| 1.6487 \| - \|
	\| 2.5641 \| 200 \| 1.631 \| - \|
	\| 2.6923 \| 210 \| 1.5296 \| - \|
	\| 2.8205 \| 220 \| 1.5704 \| - \|
	\| 2.9487 \| 230 \| 1.4634 \| - \|
	\| 3.0 \| 234 \| - \| 0.4692 \|
	\| 3.0769 \| 240 \| 1.3748 \| - \|
	\| 3.2051 \| 250 \| 1.4602 \| - \|
	\| 3.3333 \| 260 \| 1.4275 \| - \|
	\| 3.4615 \| 270 \| 1.4183 \| - \|
	\| 3.5897 \| 280 \| 1.3431 \| - \|
	\| 3.7179 \| 290 \| 1.3013 \| - \|
	\| 3.8462 \| 300 \| 1.3206 \| - \|
	\| 3.9744 \| 310 \| 1.2743 \| - \|
	\| 4.0 \| 312 \| - \| 0.4699 \|
	\| 4.1026 \| 320 \| 1.1575 \| - \|
	\| 4.2308 \| 330 \| 1.2629 \| - \|
	\| 4.3590 \| 340 \| 1.2729 \| - \|
	\| 4.4872 \| 350 \| 1.1957 \| - \|
	\| 4.6154 \| 360 \| 1.1674 \| - \|
	\| 4.7436 \| 370 \| 1.1349 \| - \|
	\| 4.8718 \| 380 \| 1.166 \| - \|
	\| 5.0 \| 390 \| 1.0891 \| 0.4707 \|
	\| 5.1282 \| 400 \| 1.0469 \| - \|
	\| 5.2564 \| 410 \| 1.124 \| - \|
	\| 5.3846 \| 420 \| 1.1325 \| - \|
	\| 5.5128 \| 430 \| 1.0691 \| - \|
	\| 5.6410 \| 440 \| 1.0255 \| - \|
	\| 5.7692 \| 450 \| 1.0164 \| - \|
	\| 5.8974 \| 460 \| 1.0451 \| - \|
	\| 6.0 \| 468 \| - \| 0.4578 \|
	\| 6.0256 \| 470 \| 0.9404 \| - \|
	\| 6.1538 \| 480 \| 1.0043 \| - \|
	\| 6.2821 \| 490 \| 0.9964 \| - \|
	\| 6.4103 \| 500 \| 1.013 \| - \|
	\| 6.5385 \| 510 \| 0.9772 \| - \|
	\| 6.6667 \| 520 \| 0.9544 \| - \|
	\| 6.7949 \| 530 \| 0.9659 \| - \|
	\| 6.9231 \| 540 \| 0.9629 \| - \|
	\| 7.0 \| 546 \| - \| 0.4576 \|
	\| 7.0513 \| 550 \| 0.8522 \| - \|
	\| 7.1795 \| 560 \| 0.9288 \| - \|
	\| 7.3077 \| 570 \| 0.9705 \| - \|
	\| 7.4359 \| 580 \| 0.9301 \| - \|
	\| 7.5641 \| 590 \| 0.9388 \| - \|
	\| 7.6923 \| 600 \| 0.8569 \| - \|
	\| 7.8205 \| 610 \| 0.9414 \| - \|
	\| 7.9487 \| 620 \| 0.8796 \| - \|
	\| 8.0 \| 624 \| - \| 0.4542 \|
	\| 8.0769 \| 630 \| 0.8504 \| - \|
	\| 8.2051 \| 640 \| 0.9054 \| - \|
	\| 8.3333 \| 650 \| 0.9035 \| - \|
	\| 8.4615 \| 660 \| 0.9167 \| - \|
	\| 8.5897 \| 670 \| 0.8546 \| - \|
	\| 8.7179 \| 680 \| 0.8508 \| - \|
	\| 8.8462 \| 690 \| 0.8945 \| - \|
	\| 8.9744 \| 700 \| 0.8676 \| - \|
	\| 9.0 \| 702 \| - \| 0.4526 \|
	\| 9.1026 \| 710 \| 0.7934 \| - \|
	\| 9.2308 \| 720 \| 0.889 \| - \|
	\| 9.3590 \| 730 \| 0.9205 \| - \|
	\| 9.4872 \| 740 \| 0.8947 \| - \|
	\| 9.6154 \| 750 \| 0.8679 \| - \|
	\| 9.7436 \| 760 \| 0.8545 \| - \|
	\| 9.8718 \| 770 \| 0.8878 \| - \|
	\| 10.0 \| 780 \| 0.8483 \| 0.4532 \|

	* The bold row denotes the saved checkpoint.

	### Framework Versions
	- Python: 3.10.11
	- Sentence Transformers: 5.1.0
	- Transformers: 4.55.2
	- PyTorch: 2.8.0+cu129
	- Accelerate: 1.10.0
	- Datasets: 3.6.0
	- Tokenizers: 0.21.4

	## 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}
	}
	```

	<!--
	## 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.
	-->