typos

pytorch_transformers/modeling_auto.py

@@ -157,7 +157,7 @@ class AutoModel(object):
             - contains `xlnet`: XLNetConfig (XLNet model)
             - contains `xlm`: XLMConfig (XLM model)
 
-            The model is set in evaluation mode by default using `model.eval()` (Dropout modules are desactivated)
+            The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
             To train the model, you should first set it back in training mode with `model.train()`
 
         Params:
@@ -179,7 +179,7 @@ class AutoModel(object):
                 - the model was saved using the `save_pretrained(save_directory)` (loaded by suppling the save directory).
             **state_dict**: an optional state dictionnary for the model to use instead of a state dictionary loaded
                 from saved weights file.
-                This option can be used if you want to create a model from a pretrained configuraton but load your own weights.
+                This option can be used if you want to create a model from a pretrained configuration but load your own weights.
                 In this case though, you should check if using `save_pretrained(dir)` and `from_pretrained(save_directory)` is not
                 a simpler option.
             **cache_dir**: (`optional`) string:

pytorch_transformers/modeling_utils.py

@@ -324,7 +324,7 @@ class PreTrainedModel(nn.Module):
     def from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs):
         r"""Instantiate a pretrained pytorch model from a pre-trained model configuration.
 
-            The model is set in evaluation mode by default using `model.eval()` (Dropout modules are desactivated)
+            The model is set in evaluation mode by default using `model.eval()` (Dropout modules are deactivated)
             To train the model, you should first set it back in training mode with `model.train()`
 
         Params:
@@ -346,7 +346,7 @@ class PreTrainedModel(nn.Module):
                 - the model was saved using the `save_pretrained(save_directory)` (loaded by suppling the save directory).
             **state_dict**: an optional state dictionnary for the model to use instead of a state dictionary loaded
                 from saved weights file.
-                This option can be used if you want to create a model from a pretrained configuraton but load your own weights.
+                This option can be used if you want to create a model from a pretrained configuration but load your own weights.
                 In this case though, you should check if using `save_pretrained(dir)` and `from_pretrained(save_directory)` is not
                 a simpler option.
             **cache_dir**: (`optional`) string:

pytorch_transformers/tokenization_bert.py

@@ -119,7 +119,7 @@ class BertTokenizer(PreTrainedTokenizer):
                 Only has an effect when do_basic_tokenize=True
             **tokenize_chinese_chars**: (`optional`) boolean (default True)
                 Whether to tokenize Chinese characters.
-                This should likely be desactivated for Japanese:
+                This should likely be deactivated for Japanese:
                 see: https://github.com/huggingface/pytorch-pretrained-BERT/issues/328
         """
         super(BertTokenizer, self).__init__(unk_token=unk_token, sep_token=sep_token,
@@ -214,7 +214,7 @@ class BasicTokenizer(object):
                 List of token not to split.
             **tokenize_chinese_chars**: (`optional`) boolean (default True)
                 Whether to tokenize Chinese characters.
-                This should likely be desactivated for Japanese:
+                This should likely be deactivated for Japanese:
                 see: https://github.com/huggingface/pytorch-pretrained-BERT/issues/328
         """
         if never_split is None: