PetarKal/qwen3-4b-EM-full-finetuned-v2
PetarKal/qwen3-4b-EM-full-finetuned-v2 is a 4 billion parameter language model, fine-tuned by PetarKal from the Qwen/Qwen3-4B base model. This model was trained using SFT (Supervised Fine-Tuning) with the TRL framework, making it suitable for general text generation tasks. It features a 32768 token context length, offering robust performance for various conversational and generative applications.
Loading preview...
Model Overview
This model, PetarKal/qwen3-4b-EM-full-finetuned-v2, is a 4 billion parameter language model developed by PetarKal. It is a fine-tuned variant of the original Qwen/Qwen3-4B base model, leveraging the TRL (Transformers Reinforcement Learning) framework for its training process.
Key Capabilities
- General Text Generation: The model is designed for a wide range of text generation tasks, as demonstrated by its quick start example involving a philosophical question.
- Supervised Fine-Tuning (SFT): Trained using SFT, indicating a focus on learning from labeled data to perform specific tasks effectively.
- Context Length: Features a substantial context window of 32768 tokens, allowing it to process and generate longer sequences of text while maintaining coherence.
Training Details
The model's training procedure utilized SFT, with the process visualized and tracked via Weights & Biases. Key framework versions used include TRL 0.29.1, Transformers 5.9.0, Pytorch 2.10.0, Datasets 4.8.5, and Tokenizers 0.22.2.
Good For
- Conversational AI: Its ability to handle complex prompts and generate coherent responses makes it suitable for chatbot development.
- Content Creation: Can be used for generating creative text, answering open-ended questions, and assisting with various writing tasks.
- Research and Experimentation: Provides a fine-tuned Qwen3-4B base for further experimentation or integration into larger systems.