google / gemma.cpp

lightweight, standalone C++ inference engine for Google's Gemma models.
Apache License 2.0
5.9k stars 499 forks source link

De-templatize Activations, add RowVectorBatch class #310

Closed copybara-service[bot] closed 1 month ago

copybara-service[bot] commented 1 month ago

De-templatize Activations, add RowVectorBatch class

Also remove most kBatchSize args.