mirror of
https://github.com/wassname/Open-Assistant.git
synced 2026-07-19 11:20:04 +08:00
[feature] add pythia and limit translation pair
This commit is contained in:
@@ -25,6 +25,10 @@ def get_tokenizer(conf):
|
||||
tokenizer.add_special_tokens({"pad_token": tokenizer.eos_token, "sep_token": "<|extratoken_100|>"})
|
||||
elif "codegen" in conf.model_name:
|
||||
tokenizer.add_special_tokens({"pad_token": "<|endoftext|>", "sep_token": "<|endoftext|>"})
|
||||
elif "pythia" in conf.model_name:
|
||||
tokenizer.add_special_tokens(
|
||||
{"pad_token": "<|padding|>", "sep_token": "<|endoftext|>", "eos_token": "<|endoftext|>"}
|
||||
)
|
||||
|
||||
additional_special_tokens = (
|
||||
[]
|
||||
|
||||
Reference in New Issue
Block a user