Some users praised OpenAI and Anthropic model cards as the best in the business, while others dismissed model cards as never useful and faulted Grok for skipping them, leaving its capabilities unknown and its prompt inadequate.
Based on 4 visible X reactions from 4 accounts; directional sample.
Ask a question below.
Published answers will appear here.
I think this is actually part of the reason nobody knows how good the Grok model is and our only familiar with it from the agent on here with the crappy system prompt. In fact, more models should have deep investigations into them and their properties just not economically incentivize to.
@krishnanrohit 😂 ok openAI and Anthropic make the best model cards in the business. I suggest that has something to do with how good their models are and how widely they are used. Not clear what the cause of relationship is
@krishnanrohit OK Mr. ignoring every thing except for first order affects directly noticeable to my face
@krishnanrohit They were never useful to begin with.
I think this is actually part of the reason nobody knows how good the Grok model is and our only familiar with it from the agent on here with the crappy system prompt. In fact, more models should have deep investigations into them and their properties just not economically incentivize to.
@krishnanrohit 😂 ok openAI and Anthropic make the best model cards in the business. I suggest that has something to do with how good their models are and how widely they are used. Not clear what the cause of relationship is
@krishnanrohit OK Mr. ignoring every thing except for first order affects directly noticeable to my face
We've had giant model cards for a year or two now. Every now and then Grok drops a frontier-ish model without one. So what have we learnt about the utility of the model card, beyond it just being interesting reading occasionally?
@deepfates I do not think the model cards were particularly useful for me to learn how useful Sonnet was, or GPT Terra is.
Some users praised OpenAI and Anthropic model cards as the best in the business, while others dismissed model cards as never useful and faulted Grok for skipping them, leaving its capabilities unknown and its prompt inadequate.
Based on 4 visible X reactions from 4 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@deepfates I do not think the model cards were particularly useful for me to learn how useful Sonnet was, or GPT Terra is.