QuantizedLinear

5 references to QuantizedLinear

Microsoft.ML.GenAI.Core (5)

Module\Attention.cs (5)

114this.o_proj = new QuantizedLinear(this._hiddenSize, this._hiddenSize, hasBias: attentionBias, dtype: dtype); 118this.qkv_proj = new QuantizedLinear(this._hiddenSize, opSize, hasBias: attentionBias, dtype: dtype); 122this.q_proj = new QuantizedLinear(this._hiddenSize, this._numHeads * this._headDim, hasBias: attentionBias, dtype: dtype); 123this.k_proj = new QuantizedLinear(this._hiddenSize, this._numKeyValueHeads * this._headDim, hasBias: attentionBias, dtype: dtype); 124this.v_proj = new QuantizedLinear(this._hiddenSize, this._numKeyValueHeads * this._headDim, hasBias: attentionBias, dtype: dtype);