Hyeongwon/P2-split4_prob_Qwen3-4B-Base_0312-01

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kPublished:Mar 13, 2026Architecture:Transformer Warm

The Hyeongwon/P2-split4_prob_Qwen3-4B-Base_0312-01 model is a 4 billion parameter language model developed by Hyeongwon, fine-tuned from the Qwen3-4B-Base architecture. This model has been trained using Supervised Fine-Tuning (SFT) with the TRL framework, focusing on enhancing its probabilistic generation capabilities. With a context length of 32768 tokens, it is designed for text generation tasks where nuanced and contextually relevant responses are crucial.

Loading preview...

Model Overview

Hyeongwon/P2-split4_prob_Qwen3-4B-Base_0312-01 is a 4 billion parameter language model, fine-tuned by Hyeongwon from the base Qwen3-4B-Base architecture. This model leverages the TRL (Transformer Reinforcement Learning) framework for its training, specifically employing Supervised Fine-Tuning (SFT).

Key Characteristics

  • Base Model: Fine-tuned from Hyeongwon/Qwen3-4B-Base.
  • Training Framework: Utilizes the TRL library for efficient fine-tuning.
  • Training Method: Trained using Supervised Fine-Tuning (SFT).
  • Context Length: Supports a substantial context window of 32768 tokens, allowing for processing and generating longer sequences of text.
  • Development Environment: Developed with specific versions of key libraries including TRL 0.25.1, Transformers 4.57.3, Pytorch 2.6.0, Datasets 3.6.0, and Tokenizers 0.22.2.

Use Cases

This model is particularly well-suited for text generation tasks that benefit from a fine-tuned base model and a large context window. Its SFT training suggests an optimization for generating coherent and contextually appropriate responses, making it suitable for applications requiring detailed and nuanced text outputs.