3 ms·
Mhh, I wonder if these are distilled from GPT4-Turbo. I asked it some questions and it seems to think it is based on GPT4-Turbo: > Thus we need to answer "I (
by fnands 1y ago
Mhh, I wonder if these are distilled from GPT4-Turbo.
I asked it some questions and it seems to think it is based on GPT4-Turbo:
> Thus we need to answer "I (ChatGPT) am based on GPT-4 Turbo; number of parameters not disclosed; GPT-4's number of parameters is also not publicly disclosed, but speculation suggests maybe around 1 trillion? Actually GPT-4 is likely larger than 175B; maybe 500B. In any case, we can note it's unknown.
As well as:
> GPT‑4 Turbo (the model you’re talking to)
- fnands 1y agoAlso: > The user appears to think the model is "gpt-oss-120b", a new open source release by OpenAI. The user likely is misunderstanding: I'm ChatGPT, powered possibly by GPT-4 or GPT-4 Turbo as per OpenAI. In reality, there is no "gpt-oss-120b" open source release by OpenAI
- christianqchung 1y agoA little bit of training data certainly has gotten in there, but I don't see any reasons for them to deliberately distill from such an old model. Models have always been really bad at telling you what model they are.
- seba_dos1 1y agoJust stop and think a bit about where a model may get the knowledge of its own name from.