NanoCode012
NanoCode012
### ⚠️ Please check that this feature request hasn't been suggested before. - [x] I searched previous [Ideas in Discussions](https://github.com/axolotl-ai-cloud/axolotl/discussions/categories/ideas) didn't find any similar feature requests. - [x] I searched...
### ⚠️ Please check that this feature request hasn't been suggested before. - [x] I searched previous [Ideas in Discussions](https://github.com/axolotl-ai-cloud/axolotl/discussions/categories/ideas) didn't find any similar feature requests. - [x] I searched...
# Description - Requires installing CCE branch https://github.com/axolotl-ai-cloud/ml-cross-entropy/pull/20 Test: - [ ] Packing (perhaps not if only in VL mode -> would need to remove from multipack array) - [...
# Description Note: - Only single GPU support at the moment. We can either go vllm route or use accelerate. TBD atm. ## Motivation and Context ## How has this...
# Description RL trainer cls plugin does not pass self.model causing `TypeError: AtroposGRPOTrainer.__init__() missing 1 required positional argument: 'model'` https://discord.com/channels/1104757954588196865/1117071926926512248/1392125907963084913 ## Motivation and Context ## How has this been tested?...
# Description Fixes #2791 Incomplete! Requires testing! The modeling code only lives on Huggingface Hub, so the mapping needs to be loaded dynamically via trust remote code. Secondly, their template...