ChesterProgrammer/V0.5.1
ChesterProgrammer/V0.5.1 is a 9 billion parameter Qwen3.5-based causal language model, finetuned by ChesterProgrammer. It was trained using LoRA with a 32768 token context length, optimized for specific personality generation from the Lucy_Personality_Complex dataset. This model is designed for applications requiring nuanced character responses and specialized conversational AI.
Loading preview...
Model Overview
ChesterProgrammer/V0.5.1 is a 9 billion parameter language model developed by ChesterProgrammer, finetuned from the DreamFast/Qwen3.5-9B-Uncensored-HauhauCS-Aggressive-Safetensor-Benchmark. This model leverages the Qwen3.5 architecture and was trained efficiently using Unsloth and Huggingface's TRL library, resulting in a 2x faster training process.
Key Training Details
The finetuning process utilized a LoRA (16-bit) method with specific parameters:
- Epochs: 2
- Batch Size: 8
- Gradient Accumulation: 1
- Learning Rate: 0.0001
- Optimizer: AdamW 8-bit
- Context Length: 2048 (during training)
- LoRA Rank: 8, Alpha: 16, Dropout: 0.05
- Dataset: Lucy_Personality_Complex (200 Samples)
Intended Use Cases
This model is particularly suited for applications requiring the generation of text with a specific personality, as it was finetuned on the Lucy_Personality_Complex dataset. Its training on a specialized dataset suggests its strength lies in nuanced character interaction and personality-driven conversational AI.