circulus/koelectra-act-v1 is a model published on Hugging Face that implements an ELECTRA architecture for multi-label classification. It is provided by Circulus Inc. and is available for direct loading in Python code. The model card shows it uses the Transformers library and PyTorch backend.
Developers load it through the standard AutoTokenizer and ElectraForMultiLabelClassification classes supplied by the Transformers package. The provided code example initializes the tokenizer and model with a single from_pretrained call that accepts an automatic device mapping. No custom training details or task-specific labels appear on the page.
The repository lists one community contribution and records 69,507 downloads in the most recent month. It has not been deployed by any of the platform's inference providers. The page contains no model card text describing intended use cases, performance metrics, or training data.
Files and versions are tracked through the standard Hugging Face interface. The model is distributed under the open terms typical of public repositories on the platform.
In the Other AI space, Koelectra Act takes a focused approach. It focuses on classifying Korean text for multiple action categories using a fine-tuned ELECTRA model. It is built as an open-source project for NLP researchers and developers working with Korean language. Koelectra Act is open source under the Open Source license. Koelectra Act is available on the web and API.
Circulus Inc. builds and maintains Koelectra Act, and the product first shipped in 2023. Key capabilities include Multi-label Classification, Transformers Compatible, and PyTorch Support. It exposes integrations via a public API.
Latest indexed changes and source events
circulus/koelectra-act-v1 verified by the PulseGate indexer
Other apps tracked under the same category.
