Update README.md

README.md

@@ -111,19 +111,19 @@ LCM (LORA) 全層適用の高速化は最大 62% になりましたが OUT 層
 結果的にこれが最適解かなと思います ※LCM 0.5 適用の場合です強度を上げればさらに高速化するはずです<br>
 (<<NON LCM STEP:16<<rev 020 LCM ALL 層 STEP:6>>LCM MID Layer STEP:12>>)<br>
 [LCM Considerations] What we learned from the rev020 test<br>
-LCM (LORA) applied to all layers resulted in a -50% speedup, but some noise remained due to the effect on the OUT layer.<br>
-Therefore, LCM (LORA) was applied only to the MID layer. The speedup was -25%, but the noise disappeared cleanly.<br> 
+LCM (LORA) applied to all layers resulted in a 62% speedup, but some noise remained due to the effect on the OUT layer.<br>
+Therefore, LCM (LORA) was applied only to the MID layer. The speedup was 25%, but the noise disappeared cleanly.<br> 
 As a result, I think this is the best solution.<br>
 ※This is the case when LCM 0.5 is applied. If the intensity is increased, the speed should be further improved.<br>
 (<<NON LCM STEP:16<<rev 020 LCM ALL Layer STEP:6>>LCM MID Layer STEP:12>>)<br>
 <br>
 [おすすめ] 高効率な LCM 利用法<br>
-LCM(LORA) をモデルにマージする場合 BM1280ch (FreeU 相当) にすると STEP 設定に余裕が生まれます<br>
+LCM(LORA) をモデルにマージする場合 bM1280ch (FreeU 相当) にすると STEP 設定に余裕が生まれます<br>
 今回のテスト版では STEP:10 以上の場合に破綻のない画像を高頻度で得られました<br>
-※BM1280ch = 1,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0<br>
+※bM1280ch = 0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0<br>
 [Recommendation] Efficient LCM usage<br>
-When merging LCM(LORA) into a model, set BM1280ch (equivalent to FreeU) to give more leeway for STEP setting.<br>
+When merging LCM(LORA) into a model, set bM1280ch (equivalent to FreeU) to give more leeway for STEP setting.<br>
 In this test version, we obtained images with no collapses at a high frequency when STEP:10 or higher was used.<br>
-※BM1280ch = 1,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0<br>
+※bM1280ch = 0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0<br>
 <br>
 <br>