update docs for tokenizer_legacy (#401)
* update docs for tokenizer_legacy * add default info
This commit is contained in:
@@ -326,6 +326,8 @@ tokenizer_type: AutoTokenizer
|
|||||||
trust_remote_code:
|
trust_remote_code:
|
||||||
# use_fast option for tokenizer loading from_pretrained, default to True
|
# use_fast option for tokenizer loading from_pretrained, default to True
|
||||||
tokenizer_use_fast:
|
tokenizer_use_fast:
|
||||||
|
# Whether to use the legacy tokenizer setting, defaults to True
|
||||||
|
tokenizer_legacy:
|
||||||
# resize the model embeddings when new tokens are added to multiples of 32
|
# resize the model embeddings when new tokens are added to multiples of 32
|
||||||
# this is reported to improve training speed on some models
|
# this is reported to improve training speed on some models
|
||||||
resize_token_embeddings_to_32x:
|
resize_token_embeddings_to_32x:
|
||||||
|
|||||||
Reference in New Issue
Block a user