1 write to o_proj
Microsoft.ML.GenAI.Core (1)
Module\Attention.cs (1)
114this.o_proj = new QuantizedLinear(this._hiddenSize, this._hiddenSize, hasBias: attentionBias, dtype: dtype);
1 reference to o_proj
Microsoft.ML.GenAI.Core (1)
Module\Attention.cs (1)
214attnOutput = this.o_proj.forward(attnOutput);