You can deploy this model and start earning money today!

bobox/
DeBERTaV3-small-ST-AdaptiveLayer-3L-ep2

Name: DeBERTaV3-small-ST-AdaptiveLayer-3L-ep2
Availability: InStock

Views

No views yet

SentenceTransformer based on microsoft/deberta-v3-small

This is a sentence-transformers model finetuned from microsoft/deberta-v3-small on the stanfordnlp/snli 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: microsoft/deberta-v3-small
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
Training Dataset:
- stanfordnlp/snli
Language: en

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': 512, 'do_lower_case': False}) with Transformer model: DebertaV2Model 
  (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})
)

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("bobox/DeBERTaV3-small-ST-AdaptiveLayer-3L-ep2")
# Run inference
sentences = [
    'These girls are having a great time looking for seashells.',
    'The girls are happy.',
    'A girl is standing outside.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

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

Evaluation

Metrics

Binary Classification

Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.6653
cosine_accuracy_threshold	0.6692
cosine_f1	0.7051
cosine_f1_threshold	0.5758
cosine_precision	0.5903
cosine_recall	0.8753
cosine_ap	0.7024
dot_accuracy	0.6308
dot_accuracy_threshold	127.0527
dot_f1	0.6984
dot_f1_threshold	101.7725
dot_precision	0.5773
dot_recall	0.8838
dot_ap	0.6558
manhattan_accuracy	0.6675
manhattan_accuracy_threshold	210.9939
manhattan_f1	0.7108
manhattan_f1_threshold	252.6531
manhattan_precision	0.6061
manhattan_recall	0.8592
manhattan_ap	0.7094
euclidean_accuracy	0.6619
euclidean_accuracy_threshold	11.2276
euclidean_f1	0.7073
euclidean_f1_threshold	12.8508
euclidean_precision	0.5879
euclidean_recall	0.8876
euclidean_ap	0.7038
max_accuracy	0.6675
max_accuracy_threshold	210.9939
max_f1	0.7108
max_f1_threshold	252.6531
max_precision	0.6061
max_recall	0.8876
max_ap	0.7094

Training Details

Training Dataset

stanfordnlp/snli

Dataset: stanfordnlp/snli at cdb5c3d
Size: 67,190 training samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 label
type string string int
details
min: 4 tokens
mean: 21.19 tokens
max: 133 tokens
min: 4 tokens
mean: 11.77 tokens
max: 49 tokens
0: 100.00%

Samples:

sentence1	sentence2	label
`Without a placebo group, we still won't know if any of the treatments are better than nothing and therefore worth giving.`	`It is necessary to use a controlled method to ensure the treatments are worthwhile.`	`0`
`It was conducted in silence.`	`It was done silently.`	`0`
`oh Lewisville any decent food in your cafeteria up there`	`Is there any decent food in your cafeteria up there in Lewisville?`	`0`

Loss: AdaptiveLayerLoss with these parameters:
json
1{ 2 "loss": "MultipleNegativesRankingLoss", 3 "n_layers_per_step": 3, 4 "last_layer_weight": 1, 5 "prior_layers_weight": 0.3, 6 "kl_div_weight": 1, 7 "kl_temperature": 1 8}

Evaluation Dataset

stanfordnlp/snli

Dataset: stanfordnlp/snli at cdb5c3d
Size: 6,626 evaluation samples
Columns: premise, hypothesis, and label
Approximate statistics based on the first 1000 samples:
premise hypothesis label
type string string int
details
min: 6 tokens
mean: 17.28 tokens
max: 59 tokens
min: 4 tokens
mean: 10.53 tokens
max: 32 tokens
0: ~48.70%
1: ~51.30%

Samples:

premise	hypothesis	label
`This church choir sings to the masses as they sing joyous songs from the book at a church.`	`The church has cracks in the ceiling.`	`0`
`This church choir sings to the masses as they sing joyous songs from the book at a church.`	`The church is filled with song.`	`1`
`A woman with a green headscarf, blue shirt and a very big grin.`	`The woman is young.`	`0`

Loss: AdaptiveLayerLoss with these parameters:
json
1{ 2 "loss": "MultipleNegativesRankingLoss", 3 "n_layers_per_step": 3, 4 "last_layer_weight": 1, 5 "prior_layers_weight": 0.3, 6 "kl_div_weight": 1, 7 "kl_temperature": 1 8}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 45
per_device_eval_batch_size: 22
learning_rate: 3e-06
weight_decay: 1e-09
num_train_epochs: 2
lr_scheduler_type: cosine
warmup_ratio: 0.5
save_safetensors: False
fp16: True
push_to_hub: True
hub_model_id: bobox/DeBERTaV3-small-ST-AdaptiveLayer-3L-ep2-n
hub_strategy: checkpoint
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 45
per_device_eval_batch_size: 22
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
learning_rate: 3e-06
weight_decay: 1e-09
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 2
max_steps: -1
lr_scheduler_type: cosine
lr_scheduler_kwargs: {}
warmup_ratio: 0.5
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: False
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: {'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
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: True
resume_from_checkpoint: None
hub_model_id: bobox/DeBERTaV3-small-ST-AdaptiveLayer-3L-ep2-n
hub_strategy: checkpoint
hub_private_repo: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
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
dispatch_batches: None
split_batches: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	max_ap
0.1004	150	4.9809	-	-
0.2001	299	-	3.8956	0.6130
0.2008	300	3.8459	-	-
0.3012	450	3.1941	-	-
0.4003	598	-	3.2066	0.6526
0.4016	600	2.7939	-	-
0.5020	750	2.3082	-	-
0.6004	897	-	2.4595	0.6884
0.6024	900	1.9658	-	-
0.7028	1050	1.6975	-	-
0.8005	1196	-	2.0292	0.7010
0.8032	1200	1.528	-	-
0.9036	1350	1.3763	-	-
1.0007	1495	-	1.8192	0.7071
1.0040	1500	1.262	-	-
1.1044	1650	1.2033	-	-
1.2008	1794	-	1.6673	0.7082
1.2048	1800	1.1221	-	-
1.3052	1950	1.0963	-	-
1.4009	2093	-	1.5816	0.7103
1.4056	2100	1.0742	-	-
1.5060	2250	1.0242	-	-
1.6011	2392	-	1.5368	0.7094
1.6064	2400	1.0036	-	-
1.7068	2550	1.0143	-	-
1.8012	2691	-	1.5158	0.7094
1.8072	2700	0.9799	-	-
1.9076	2850	0.9777	-	-

Framework Versions

Python: 3.10.13
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.1.2
Accelerate: 0.30.1
Datasets: 2.19.2
Tokenizers: 0.19.1

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",
}

AdaptiveLayerLoss

bibtex

@misc{li20242d,
    title={2D Matryoshka Sentence Embeddings}, 
    author={Xianming Li and Zongxi Li and Jing Li and Haoran Xie and Qing Li},
    year={2024},
    eprint={2402.14776},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

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}
}

	sentence1	sentence2	label
type	string	string	int
details	min: 4 tokens mean: 21.19 tokens max: 133 tokens	min: 4 tokens mean: 11.77 tokens max: 49 tokens	0: 100.00%