barc0/Llama-3.1-ARC-Potpourri-Induction-8B

barc0/Llama-3.1-ARC-Potpourri-Induction-8B is an 8 billion parameter instruction-tuned model based on Meta-Llama-3.1-8B-Instruct, fine-tuned specifically for advanced pattern recognition and inductive reasoning tasks. It excels at solving complex puzzles, particularly those involving grid transformations and logical deduction, by generating Python solutions. The model leverages a 32768-token context length and is optimized for tasks requiring deep analytical capabilities.