How does this compare with NeuralHermes 2.5?

#3
by tarruda - opened

https://huggingface.co/mlabonne/NeuralHermes-2.5-Mistral-7B description: "NeuralHermes is based on the teknium/OpenHermes-2.5-Mistral-7B model that has been further fine-tuned with Direct Preference Optimization (DPO) using the mlabonne/chatml_dpo_pairs dataset"

How does this differ from neuralhermes (other than the dataset) ?

@mlabonne

Shouldn't be too different but this model is probably slightly better than NeuralHermes (haven't tested it, but I would choose this one by default).

Is it multi-lingual as it is finetuned from Mistral ?

Sign up or log in to comment