3 ms·
The "Tuned" version is finetuned on the output of ChatGPT, which is why it is non-commercial licensed. The "Base" version does not do this (and is commercial l
by MacsHeadroom 4y ago
The "Tuned" version is finetuned on the output of ChatGPT, which is why it is non-commercial licensed.
The "Base" version does not do this (and is commercial licensed).
- ftxbro 4y agoThanks! It's sad that this is the answer, but I understand. I wish sometime we can access the base models of GPT-4 without the 'As an AI language model' guardrails!
- MacsHeadroom 4y agoThat's not going to happen. But it's likely that StableLM 175B will rival GPT-4. Also, you can finetune Base StableLM yourself on any consumer GPU with 8GB of VRAM in a couple of hours and it will be commercial licensed. (using https://github.com/johnsmith0031/alpaca_lora_4bit https://github.com/johnsmith0031/alpaca_lora_4bit) You can even use the exact same dataset StabilityAI used. (Although there are better ones, with more GPT-4 data.)