mirror of https://github.com/google/gemma.cpp.git
5a05857deb
* Adds and uses a new `AttentionActivationPtrs` that holds non-owning `MatPtrs`. Acts as a view into `AttentionActivations`. * Updates `QBatch` to hold non-owning `MatPtr`s to the kv caches. * Enables the `MatPtrT` default constructor for simpler initializations. * Pulls out and passes `LayerWeightsPtrs::query_norm_scale` directly. While `LayerWeightsPtrs` already held non-owning `MatPtr`s, this change avoids the need to find and construct several empty weight tensors just to construct one `query_norm_scale` tensor. PiperOrigin-RevId: 824584177 |
||
|---|---|---|
| .. | ||
| allocator.cc | ||
| allocator.h | ||
| args.h | ||
| basics.cc | ||
| basics.h | ||
| basics_test.cc | ||
| mat.cc | ||
| mat.h | ||
| test_util.h | ||
| threading.cc | ||
| threading.h | ||
| threading_context.cc | ||
| threading_context.h | ||
| threading_test.cc | ||
| topology.cc | ||
| topology.h | ||
| zones.cc | ||
| zones.h | ||