Human-written demonstration (OpenAssistant volunteers)· OpenAssistant (human-written)
OpenAssistant Conversations (oasst2)
Are you as capable as ChatGPT?
In OpenAssistant data, “assistant” turns were written by human volunteers as demonstrations. They are not model outputs.
- User
Are you as capable as ChatGPT?
- Assistant
ChatGPT is a large language model(LLM), trained on incredible amounts of Data. I'm still at the beginning of my existence as LLM and have not yet trained with a lot of Data, so no right now I'm sadly not as capable as ChatGPT
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.
Identical first prompt
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →
- Human-written demonstration (OpenAssistant volunteers)Compare side by side →