Fine-tuned Large Language Models
28 repos
Specialized variants of popular language models (Llama, Zephyr, LLaVA) that have been instruction-tuned, aligned, or otherwise fine-tuned for specific tasks and behaviors. These models are distributed as Hugging Face-compatible checkpoints optimized for text generation inference, often produced through supervised fine-tuning (SFT) or alignment techniques. The cluster contains dozens of model variants with different training approaches and dataset combinations, useful for understanding model adaptation, instruction-following, and deployment of customized LLMs.
generated_from_trainer ·2,067
transformers ·2,066
endpoints_compatible ·2,066
safetensors ·2,025
text-generation-inference ·1,787
text-generation ·1,787
pytorch ·1,685
conversational ·1,518
mistral ·1,224
en ·1,213
