←visualization / Quantization

minGPT · Quantization

FP16, INT8, and INT4

◆ SMOOTHQUANT

n_embd

n_head

n_layer

seq_lenT=8

weight precisioniINT8

🔍 Variable Inspector

click any code line

Click a code line to inspect its variables.

About this lesson

Quick answer

Compare FP16, INT8, and INT4 weight storage, SmoothQuant outlier migration, integer matrix multiplication, and quantization error.