- 07 Oct, 2023 2 commits
-
-
Casper authored
Only apply attention mask if seqlen is greater than 1
-
Casper Hansen authored
-
- 06 Oct, 2023 10 commits
-
-
Casper authored
-
Casper authored
Refactor cache and embedding modules
-
Casper Hansen authored
-
Casper Hansen authored
-
Casper Hansen authored
-
Casper Hansen authored
-
Casper Hansen authored
-
Casper Hansen authored
-
Casper Hansen authored
-
Casper Hansen authored
-
- 05 Oct, 2023 7 commits
- 03 Oct, 2023 2 commits
-
-
Casper authored
-
Casper Hansen authored
-
- 02 Oct, 2023 8 commits
-
-
Casper Hansen authored
-
Casper authored
Mistral fused modules
-
Casper Hansen authored
-
Casper Hansen authored
-
Casper authored
Fix Falcon n_kv_heads parameter
-
Casper authored
Fix unexpected keyword
-
Casper Hansen authored
-
Casper Hansen authored
-
- 01 Oct, 2023 2 commits
- 27 Sep, 2023 9 commits
-
-
Casper authored
-
Casper authored
-
Casper authored
Offloading to cpu and disk
-
Casper authored
Add low_cpu_mem_usage=True in example
-
Casper Hansen authored
-
Casper authored
Mistral support
-
Casper Hansen authored
-
Casper Hansen authored
-
s4rduk4r authored
-