nbeerbower/mistral-nemo-kartoffel-12B
nbeerbower/mistral-nemo-kartoffel-12B is a 12 billion parameter causal language model developed by nbeerbower, fine-tuned from Mahou-1.5-mistral-nemo-12B-lorablated. This model utilizes ORPO tuning with QLoRA configuration, trained for 2 epochs on various datasets. It is designed for general language generation tasks, leveraging its 32768 token context length for comprehensive understanding and response generation.
Loading preview...
Overview
nbeerbower/mistral-nemo-kartoffel-12B is a 12 billion parameter language model, building upon the Mahou-1.5-mistral-nemo-12B-lorablated base model. It has been further fine-tuned using the ORPO (Optimized Reward Policy Optimization) method over two epochs, leveraging 8x A100 GPUs for training. The model incorporates a QLoRA configuration, which includes 4-bit quantization and double quantization for efficient memory usage during training and inference.
Key Capabilities
- Causal Language Modeling: Designed for generating coherent and contextually relevant text.
- Efficient Fine-tuning: Utilizes QLoRA with specific
target_modulesfor effective parameter-efficient fine-tuning. - ORPO Tuning: Benefits from ORPO, a method aimed at aligning model outputs with desired behaviors and preferences.
Good for
- General Text Generation: Suitable for a wide range of language generation tasks where a 12B parameter model is appropriate.
- Research and Experimentation: Provides a fine-tuned model for exploring the effects of ORPO and QLoRA on a Mistral-Nemo base.
- Applications requiring a 32K Context Window: Its 32768 token context length allows for processing and generating longer sequences of text.
Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.