2 writes to _cache
Microsoft.ML.GenAI.LLaMA (2)
Module\LlamaModel.cs (2)
47
this.
_cache
= new DynamicKVCache();
62
this.
_cache
= input.OverrideCache;
2 references to _cache
Microsoft.ML.GenAI.LLaMA (2)
Module\LlamaModel.cs (2)
135
pastKeyValue: this.
_cache
,
152
return new CausalLMModelOutput(lastHiddenState: hiddenStates, allHiddenStates: allHiddenStates.ToArray(), attentions: allAttentions.ToArray(), cache: this.
_cache
);