Undi95/LewdMistral-7B-0.2
Undi95/LewdMistral-7B-0.2 is a 7 billion parameter language model, fine-tuned from Mistral-7B-v0.2, specifically optimized for roleplay tasks. This model was developed by Undi95 through a full fine-tuning process on multiple roleplay datasets. It serves as an experimental base for merging with other Mistral 0.1 finetunes, demonstrating its utility in creating specialized models like BigL-7B.
Loading preview...
Model Overview
Undi95/LewdMistral-7B-0.2 is a 7 billion parameter language model, derived from a full fine-tune of the Mistral-7B-v0.2 architecture. This model was developed by Undi95 with a specific focus on roleplay (RP) tasks, utilizing multiple RP datasets for its training.
Key Characteristics
- Base Model: Fine-tuned from Mistral-7B-v0.2.
- Parameter Count: 7 billion parameters.
- Context Length: Supports a context length of 4096 tokens.
- Training: Underwent a full fine-tuning process over 2 epochs on specialized RP datasets.
- Experimental Purpose: Primarily created as an experiment to assess the feasibility of merging new data from Mistral 0.2 into existing 0.1 finetunes.
Use Cases
- Roleplay Applications: Optimized for generating responses in roleplay scenarios.
- Model Merging: Suitable as a base for further training or merging with other Mistral 0.1 finetuned models, as demonstrated by its use in creating BigL-7B.
- Research & Development: Useful for developers experimenting with finetuning and merging Mistral-based models for niche applications.
Prompt Format
This model utilizes the Alpaca prompt template for instruction-based interactions.