Update README.md

README.md

@@ -123,13 +123,13 @@ Strongest Imatrix effect(s) are IQ quants and the strength of the effect is inve
 DI-Matrix and TRI-Matrix are "averages" of 2 and 3 imatrix datasets (generated specifically for a model, separately). This averaging
 can "trim" some effects and/or add some "traits" and make better quants.
 
-In the case of abliterated model(s), I find imatrix quants can fix minor issues caused by the abliteration process in some cases.
+In the case of abliterated model(s), I find "imatrixing" quants can fix minor issues caused by the abliteration process in some cases.
 
 Depending on your use case(s) regular imatrix, and/or DI/TRI imatrix quant(s) may meet different use case(s) requirement(s).
 
 To test: Try 2-5 generations per quant (same prompt, exact same settings), then evaluate output/thinking.
 
-The Imatrix effect itself depends on the model being imatrixed, strength of the imatrix dataset and the quant(s) targeted.
+The Imatrix effect itself depends on the model being imatrixed, strength of the imatrix dataset(s) and the quant(s) targeted.
 
 The Q8 quants (only) have been modified to allow limited imatrix effect(s) in this case: the output tensor only.