new paper?
#1
by
Yhyu13
- opened
Hi,
Would you release paper for v2 of transNormerLLM?
I would like to see some comparisons with Mamba and pythia for 2.8/3B size, would it be possible?
Thanks!
Hello and thank you for your interest in TransNormerLLM2!
We compared the architecture diffierence in diff-of-transnormerllm2. Additionally, we're planning to update our benchmark results for the 3B size model soon. These updates will be available in the section "benchmark-results" at this link: Benchmark Results for TransNormerLLM2-3B-300B.