About EVA Qwen2.5 14B
A model specializing in RP and creative writing, this model is based on Qwen2.5-14B, fine-tuned with a mixture of synthetic and natural data.
It is trained on 1.5M tokens of role-play data, and fine-tuned on 1.5M tokens of synthetic data.
Specifications
- Provider
- Eva unit 01
- Context Length
- 32,768 tokens
- Input Types
- text
- Output Types
- text
- Category
- Qwen
- Added
- 9/30/2024