"lm_eval/models/seq2seq.py" did not exist on "4e9412d51ee7adf5cb89303f7c88ac3fbb7b0703"
- 07 Jun, 2023 5 commits
-
-
Alan Turner authored
-
Alan Turner authored
-
Alan Turner authored
-
Alan Turner authored
-
Alan Turner authored
-
- 06 Jun, 2023 5 commits
-
-
Alan Turner authored
-
Alan Turner authored
-
Alan Turner authored
-
Umang Yadav authored
-
Umang Yadav authored
Sigmoid approximation for GeLU was introduced in #1299 for Fp16. The sigmoid approximation is known to get better perf but lower accuracy. https://arxiv.org/pdf/1606.08415.pdf
-
- 02 Jun, 2023 5 commits
- 01 Jun, 2023 6 commits
- 31 May, 2023 7 commits
- 30 May, 2023 9 commits
- 26 May, 2023 3 commits
