Harry Mellor b245d8e73f Apply logit softcapping in Transformers modelling backend (#52173) ...

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

2026-08-13 16:21:55 +00:00

..

fusers

Hardware-agnostic model definition via HF transformer backend (1/N) (#49458)

2026-08-13 11:21:50 +01:00

__init__.py

Support MLA properly in the Transformers modeling backend (#48250)

2026-08-04 13:24:11 +01:00

base.py

Fix docs on main (#51773)

2026-08-10 23:52:04 -07:00

causal.py

Apply logit softcapping in Transformers modelling backend (#52173)

2026-08-13 16:21:55 +00:00

fuser.py

Support MLA properly in the Transformers modeling backend (#48250)

2026-08-04 13:24:11 +01:00

fx_utils.py

Support MLA properly in the Transformers modeling backend (#48250)

2026-08-04 13:24:11 +01:00

layers.py

Hardware-agnostic model definition via HF transformer backend (1/N) (#49458)

2026-08-13 11:21:50 +01:00

legacy.py

[Bugfix] Fix RoBERTa position_ids accumulation on CUDA graph padding (#37884)

2026-03-23 15:15:12 +00:00

moe.py

[MoE Refactor] Rename FusedMoE to FusedMoEFactory (#44941)

2026-07-31 11:28:18 +01:00

multimodal.py

[ROCm] Remove stale SDPA and skinny GEMM workarounds (#50907)

2026-08-11 14:19:16 -04:00

pooling.py

[Doc] Fix duplicate words in comments (#36713)

2026-03-10 21:28:31 -07:00

utils.py

Fully generalise input embedding handling in Transformers modelling backend (#51247)

2026-08-06 16:21:24 +00:00