Do things that I am not proud of

This commit is contained in:
oobabooga 2023-04-09 23:40:17 -03:00
parent 65552d2157
commit dba2000d2b

View File

@ -179,9 +179,14 @@ def load_model(model_name):
tokenizer = AutoTokenizer.from_pretrained(Path(f"{shared.args.model_dir}/gpt-j-6B/")) tokenizer = AutoTokenizer.from_pretrained(Path(f"{shared.args.model_dir}/gpt-j-6B/"))
elif type(model) is transformers.LlamaForCausalLM: elif type(model) is transformers.LlamaForCausalLM:
tokenizer = LlamaTokenizer.from_pretrained(Path(f"{shared.args.model_dir}/{shared.model_name}/"), clean_up_tokenization_spaces=True) tokenizer = LlamaTokenizer.from_pretrained(Path(f"{shared.args.model_dir}/{shared.model_name}/"), clean_up_tokenization_spaces=True)
tokenizer.eos_token_id = 2 # Leaving this here until the LLaMA tokenizer gets figured out.
tokenizer.bos_token_id = 1 # For some people this fixes things, for others it causes an error.
tokenizer.pad_token_id = 0 try:
tokenizer.eos_token_id = 2
tokenizer.bos_token_id = 1
tokenizer.pad_token_id = 0
except:
continue
else: else:
tokenizer = AutoTokenizer.from_pretrained(Path(f"{shared.args.model_dir}/{shared.model_name}/")) tokenizer = AutoTokenizer.from_pretrained(Path(f"{shared.args.model_dir}/{shared.model_name}/"))
tokenizer.truncation_side = 'left' tokenizer.truncation_side = 'left'