This is a SetFit model that can be used for Text Classification. This SetFit model uses jhu-clsp/mmBERT-base as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification.

The model has been trained using an efficient few-shot learning technique that involves:

Fine-tuning a Sentence Transformer with contrastive learning.
Training a classification head with features from the fine-tuned Sentence Transformer.

Model Details

Model Description

Model Type: SetFit
Sentence Transformer body: jhu-clsp/mmBERT-base
Classification head: a LogisticRegression instance
Maximum Sequence Length: 8192 tokens
Number of Classes: 2 classes

Model Sources

Repository: SetFit on GitHub
Paper: Efficient Few-Shot Learning Without Prompts
Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts |

Evaluation

Metrics

Label	Accuracy
all	0.9992

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

python

from setfit import SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("johnpaulbin/toxicity-setfit-4-large")
# Run inference
preds = model("add me for gifts")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	1	4.5954	81

Label	Training Sample Count
not toxic	8589
toxic	4455

Training Hyperparameters

batch_size: (128, 128)
num_epochs: (1, 1)
max_steps: -1
sampling_strategy: num_iterations
num_iterations: 8
body_learning_rate: (2e-05, 1e-05)
head_learning_rate: 0.01
loss: CosineSimilarityLoss
distance_metric: cosine_distance
margin: 0.25
end_to_end: False
use_amp: True
warmup_proportion: 0.1
l2_weight: 0.01
seed: 42
eval_max_steps: -1
load_best_model_at_end: True

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0006	1	0.3232	-
0.0307	50	0.2699	-
0.0613	100	0.2104	-
0.0920	150	0.113	-
0.1226	200	0.0829	-
0.1533	250	0.0521	-
0.1839	300	0.0439	-
0.2146	350	0.0314	-
0.2452	400	0.0292	-
0.2759	450	0.0215	-
0.3066	500	0.018	-
0.3372	550	0.0172	-
0.3679	600	0.0119	-
0.3985	650	0.0113	-
0.4292	700	0.0082	-
0.4598	750	0.0098	-
0.4905	800	0.0075	-
0.5212	850	0.0073	-
0.5518	900	0.0061	-
0.5825	950	0.0052	-
0.6131	1000	0.0044	-
0.6438	1050	0.0058	-
0.6744	1100	0.0055	-
0.7051	1150	0.0054	-
0.7357	1200	0.0041	-
0.7664	1250	0.0049	-
0.7971	1300	0.0057	-
0.8277	1350	0.004	-
0.8584	1400	0.004	-
0.8890	1450	0.0039	-
0.9197	1500	0.0036	-
0.9503	1550	0.0038	-
0.9810	1600	0.0037	-
1.0	1631	-	0.0021

Framework Versions

Python: 3.12.12
SetFit: 1.2.0.dev0
Sentence Transformers: 5.2.0
Transformers: 4.57.3
PyTorch: 2.9.0+cu126
Datasets: 4.0.0
Tokenizers: 0.22.2

Citation

BibTeX

bibtex

@article{https://doi.org/10.48550/arxiv.2209.11055,
    doi = {10.48550/ARXIV.2209.11055},
    url = {https://arxiv.org/abs/2209.11055},
    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
    title = {Efficient Few-Shot Learning Without Prompts},
    publisher = {arXiv},
    year = {2022},
    copyright = {Creative Commons Attribution 4.0 International}
}