Update README.md
Browse files
README.md
CHANGED
|
@@ -31,10 +31,10 @@ Although llama.cpp can be used to reduce the size of the file with various quant
|
|
| 31 |
|
| 32 |
以下はllama.cppの作者(ggerganov)による推奨パラメーターです
|
| 33 |
|
| 34 |
-
-e (改行\nをエスケープ)
|
| 35 |
-
--temp 0 (最も確率の高いトークンのみを選択)
|
| 36 |
-
--repeat-penalty 1.0 (繰り返しペナルティをオフ)
|
| 37 |
-
--no-penalize-nl
|
| 38 |
|
| 39 |
- Temperature (--temp): Lowering this value will make the model more likely to select more confident (i.e., more common) words.
|
| 40 |
- Top P (--top_p): Setting this value even lower will narrow the range of words considered by the model and produce more consistent text.
|
|
|
|
| 31 |
|
| 32 |
以下はllama.cppの作者(ggerganov)による推奨パラメーターです
|
| 33 |
|
| 34 |
+
- -e (改行\nをエスケープ)
|
| 35 |
+
- --temp 0 (最も確率の高いトークンのみを選択)
|
| 36 |
+
- --repeat-penalty 1.0 (繰り返しペナルティをオフ)
|
| 37 |
+
- --no-penalize-nl (改行にペナルティをあたえない)
|
| 38 |
|
| 39 |
- Temperature (--temp): Lowering this value will make the model more likely to select more confident (i.e., more common) words.
|
| 40 |
- Top P (--top_p): Setting this value even lower will narrow the range of words considered by the model and produce more consistent text.
|