Commit Graph
6 Commits
Author SHA1 Message Date
chenyuandGitHub 630edcffd8 remove .float calls in olmoe (#11610)
still matches torch
2025-08-10 20:33:22 -04:00
chenyuandGitHub 3eba3d6ee9 don't pass model in convert_from_huggingface and convert_from_gguf (#10094)
it only needs n_layers
2025-04-28 20:11:19 -04:00
geohot 865f23dd7b olmoe memory usage cleanups 2025-03-19 12:28:18 +08:00
chenyuandGitHub 1ea4876dfa olmoe touchups (#9499)
GlobalCounters.reset() and only validate if temperature is 0
2025-03-18 15:25:45 -04:00
geohotstanandGitHub f7506c6c25 JIT OLMoE (#9396)
* jit the forward

* might timeout, idk just send it

* this is dumb

* naive bitonic lol

* idk if this is correct, but that squeeze before is definitly not

* vectorized bitonic sort, but still slow

* yay 1 layer is correct

* alright its pretty good

* good enough

* rerun CI

* nit improve comment
2025-03-18 14:49:02 -04:00
George HotzandGitHub 25847080f0 olmoe (from stream, wip) (#9390)
* olmoest working (but not)

* it's correct

* compare ropes

* old code wasn't wrong

* default device

* no metal

* fix permute

* working

* more minimal
2025-03-10 13:46:33 +08:00