Benchmark seems confused in terms of Llama-70B VS 8B
#1
by
Blackroot
- opened
The benchmark shows what I'm assuming is Llama-3-70B-Instruct, but also shows "8B" at the bottom, which is quite confusing. Just to avoid any doubt, could you clarify which model of the Llama-3 was actually benchmarked here?
Cheers for the model!
It scores 80 on MMLU...clearly Llama 3 70B
Thanks for your feedback! It is a typo and we have fixed it.
nguyenbh
changed discussion status to
closed