This is a sentence-transformers model finetuned from 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
Maximum Sequence Length: 8192 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- train

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': 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:

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_55_50")
# 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([[0.2540, 0.0537, 0.0780]])

Evaluation

Metrics

Information Retrieval

Dataset: Snowflake/snowflake-arctic-embed-m-v2.0
Evaluated with scripts.InformationRetrievalEvaluatorCustom.InformationRetrievalEvaluatorCustom with these parameters:
json
1{ 2 "query_prompt_name": "query", 3 "corpus_prompt_name": "query" 4}

Metric	Value
cosine_accuracy@1	0.1898
cosine_accuracy@3	0.7153
cosine_accuracy@5	0.8029
cosine_accuracy@10	0.854
cosine_precision@1	0.1898
cosine_precision@3	0.4428
cosine_precision@5	0.4657
cosine_precision@10	0.446
cosine_recall@1	0.0083
cosine_recall@3	0.0989
cosine_recall@5	0.1659
cosine_recall@10	0.237
cosine_ndcg@10	0.459
cosine_mrr@10	0.4677
cosine_map@100	0.2944

Training Details

Training Dataset

train

Dataset: train
Size: 19,964 training samples
Columns: query and answer
Approximate statistics based on the first 1000 samples:
query answer
type string string
details
min: 4 tokens
mean: 27.77 tokens
max: 615 tokens
min: 3 tokens
mean: 22.87 tokens
max: 151 tokens

Samples:

query	answer
`Wie kann man die Jahresurlaubsübersicht exportieren?`	`über das 3 Punkte Menü rechts oben. Mitarbeiter auswählen und exportieren`
`1. Vertragsabschlüsse werden nicht übertragen`
2. Kinder kommen nicht von nach
3. Absage kann bei Portalstatus nicht erstellt werden.	`Ticket`
Userin gebeten sich an den Support zu wenden, da der Fehler liegt.
`Wird im Anmeldeportal nicht gefunden.`	`Die Schnittstelle war noch nicht aktiviert und Profil ebenfalls nicht.`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false 5}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 32
per_device_eval_batch_size: 64
gradient_accumulation_steps: 8
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

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 32
per_device_eval_batch_size: 64
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 8
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: {}

Training Logs

Epoch	Step	Training Loss	Snowflake/snowflake-arctic-embed-m-v2.0_cosine_ndcg@10
0.1282	10	2.8279	-
0.2564	20	2.7011	-
0.3846	30	2.6182	-
0.5128	40	2.3893	-
0.6410	50	2.2499	-
0.7692	60	2.1048	-
0.8974	70	1.987	-
1.0	78	-	0.5043
1.0256	80	1.7766	-
1.1538	90	1.7516	-
1.2821	100	1.6332	-
1.4103	110	1.5975	-
1.5385	120	1.5437	-
1.6667	130	1.4739	-
1.7949	140	1.3988	-
1.9231	150	1.3845	-
2.0	156	-	0.4853
2.0513	160	1.2183	-
2.1795	170	1.2841	-
2.3077	180	1.2558	-
2.4359	190	1.2305	-
2.5641	200	1.2234	-
2.6923	210	1.1089	-
2.8205	220	1.1591	-
2.9487	230	1.0641	-
3.0	234	-	0.4735
3.0769	240	1.0085	-
3.2051	250	1.0507	-
3.3333	260	1.0183	-
3.4615	270	1.0208	-
3.5897	280	0.9587	-
3.7179	290	0.9273	-
3.8462	300	0.9171	-
3.9744	310	0.9076	-
4.0	312	-	0.4704
4.1026	320	0.8029	-
4.2308	330	0.8903	-
4.3590	340	0.8794	-
4.4872	350	0.851	-
4.6154	360	0.823	-
4.7436	370	0.7819	-
4.8718	380	0.7974	-
5.0	390	0.7552	0.4693
5.1282	400	0.7336	-
5.2564	410	0.7652	-
5.3846	420	0.7597	-
5.5128	430	0.7481	-
5.6410	440	0.6982	-
5.7692	450	0.6817	-
5.8974	460	0.7136	-
6.0	468	-	0.4652
6.0256	470	0.6233	-
6.1538	480	0.6739	-
6.2821	490	0.6646	-
6.4103	500	0.6614	-
6.5385	510	0.6699	-
6.6667	520	0.6291	-
6.7949	530	0.6344	-
6.9231	540	0.6459	-
7.0	546	-	0.4635
7.0513	550	0.5652	-
7.1795	560	0.6227	-
7.3077	570	0.6308	-
7.4359	580	0.6253	-
7.5641	590	0.6315	-
7.6923	600	0.5571	-
7.8205	610	0.6234	-
7.9487	620	0.5742	-
8.0	624	-	0.4611
8.0769	630	0.5583	-
8.2051	640	0.5817	-
8.3333	650	0.5913	-
8.4615	660	0.6025	-
8.5897	670	0.5726	-
8.7179	680	0.5492	-
8.8462	690	0.5907	-
8.9744	700	0.5756	-
9.0	702	-	0.4606
9.1026	710	0.5134	-
9.2308	720	0.5861	-
9.3590	730	0.6	-
9.4872	740	0.5839	-
9.6154	750	0.5688	-
9.7436	760	0.5443	-
9.8718	770	0.5687	-
10.0	780	0.5608	0.4590

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}
}

	query	answer
type	string	string
details	min: 4 tokens mean: 27.77 tokens max: 615 tokens	min: 3 tokens mean: 22.87 tokens max: 151 tokens