readme

README.md

@@ -38,27 +38,27 @@ widely used by the language modeling community) is given as the default.
 ### Model
 Here are some example scripts.
 
-Large character-level model (`LSTM-CharCNN-Large' in the paper).
+Large character-level model (`LSTM-CharCNN-Large` in the paper).
 This is the default: should get ~82 on valid and ~79 on test.
 ```
 th main.lua -gpuid 0 -savefile char-large
 ```
 
-Small character-level model ('LSTM-CharCNN-Small' in the paper).
+Small character-level model (`LSTM-CharCNN-Small` in the paper).
 This should get ~96 on valid and ~93 on test.
 ```
 th main.lua -gpuid 0 -savefile char-small -rnn_size 300 -highway_layers 1 
 -kernels '{1,2,3,4,5,6}' -feature_maps '{25,50,75,100,125,150}'
 ```
 
-Large word-level model.
+Large word-level model (`LSTM-Word-Large` in the paper).
 This should get ~89 on valid and ~85 on test.
 ```
 th main.lua -gpuid 0 -savefile word-large -word_vec_size 650 -highway_layers 0 
 -use_chars 0 -use_words 1
 ```
 
-Small word-level model.
+Small word-level model (`LSTM-Word-Small` in the paper).
 This should get ~101 on valid and ~98 on test.
 ```
 th main.lua -gpuid 0 -savefile word-small -word_vec_size 200 -highway_layers 0