Firstly, this paper is really really impression. This resolves one of main issues của currently LLMs.
But I have many confused, maybe because I don't really welling in this field. So, I have a question for my case: I want to ft phi-3-mini-4k-instruct use my own datasets at 4k, then I want to extends context length to 256k.
Can you give me a instructions or code example ? Finally, many thanks for all.
Firstly, this paper is really really impression. This resolves one of main issues của currently LLMs.
But I have many confused, maybe because I don't really welling in this field. So, I have a question for my case: I want to ft
phi-3-mini-4k-instructuse my own datasets at 4k, then I want to extends context length to 256k.Can you give me a instructions or code example ? Finally, many thanks for all.