I mean this should be expected, models learn from us, and "WE" game the metrics time after time. I also don't think it will just disappear just because we clean pretraining data. I believe the deeper reason is…
Most people use divs for everything now. This is a breath of fresh air.
I understand the appeal due to the speed
I feel like, since 4.0, it is pretty much the same model but with new names. They are just improving the CoT and function calling.
[dead]
I mean this should be expected, models learn from us, and "WE" game the metrics time after time. I also don't think it will just disappear just because we clean pretraining data. I believe the deeper reason is…
Most people use divs for everything now. This is a breath of fresh air.
I understand the appeal due to the speed
I feel like, since 4.0, it is pretty much the same model but with new names. They are just improving the CoT and function calling.
[dead]