nbeerbower/mistral-nemo-cc-12B
nbeerbower/mistral-nemo-cc-12B is a 12 billion parameter experimental language model developed by nbeerbower, fine-tuned from nbeerbower/mistral-nemo-gutenberg-12B-v3 on the flammenai/casual-conversation-DPO dataset. This model is specifically optimized for generating conversational text, formatting dialogue sequentially using ChatML. Its primary use case is for applications requiring casual, sequential conversation generation.
Loading preview...
Overview
nbeerbower/mistral-nemo-cc-12B is a 12 billion parameter experimental language model developed by nbeerbower. It is a fine-tuned version of nbeerbower/mistral-nemo-gutenberg-12B-v3, specifically adapted using the flammenai/casual-conversation-DPO dataset. The model was fine-tuned for 3 epochs on an A100 GPU via Google Colab, with a focus on formatting conversation data sequentially using ChatML.
Key Capabilities
- Casual Conversation Generation: Optimized for producing natural, sequential dialogue.
- ChatML Formatting: Designed to format conversational output according to the ChatML standard.
- Experimental Fine-tune: Represents an exploration into DPO fine-tuning for conversational datasets.
Good for
- Dialogue Systems: Ideal for chatbots or applications requiring human-like conversational flow.
- Research in Conversational AI: Useful for experimenting with DPO fine-tuning on dialogue datasets.
- Sequential Text Generation: Suited for tasks where turn-based, formatted conversation is required.