-
Notifications
You must be signed in to change notification settings - Fork 35
Issues
is:issue state:open
is:issue state:open
Issue creation is restricted in this repository
Search results
- Status: Open.#379 In explosion/curated-transformers;
Truncation of sequences that are beyond the model's maximum length
feat/tokenizationFeature: Tokenization/piecerFeature: Tokenization/piecertype/bugType: BugType: Bugtype/featureType: FeatureType: FeatureStatus: Open.#359 In explosion/curated-transformers;Add suggested PyTorch LLM optimizations
feat/generationFeature: GenerationFeature: Generationfeat/modelFeature: modelsFeature: modelsStatus: Open.#356 In explosion/curated-transformers;Move the old Falcon architecuture to the extras/addons pacakage
type/maintenanceType: MaintenanceType: MaintenanceStatus: Open.Add support for attention sinks
feat/layersFeature: LayersFeature: Layersfeat/modelFeature: modelsFeature: modelstype/featureType: FeatureType: FeatureStatus: Open.Support DeBERTa v2/3
feat/modelFeature: modelsFeature: modelstype/featureType: FeatureType: FeatureStatus: Open.Add a an extras/contrib package
type/maintenanceType: MaintenanceType: MaintenanceStatus: Open.Expose more outputs through the
Generatorinterfacefeat/generationFeature: GenerationFeature: Generationtype/featureType: FeatureType: FeatureStatus: Open.Make
QkvModeADT-likefeat/layersFeature: LayersFeature: Layerstype/maintenanceType: MaintenanceType: MaintenanceStatus: Open.Convert QKV projection splitting methods into Torch modules
feat/layersFeature: LayersFeature: Layerstype/maintenanceType: MaintenanceType: MaintenanceStatus: Open.Option to only return the last hidden layer output from models
feat/modelFeature: modelsFeature: modelstype/featureType: FeatureType: FeatureStatus: Open.Add support for Mistral
feat/modelFeature: modelsFeature: modelstype/featureType: FeatureType: FeatureStatus: Open.