Temporarily set the `instruct` version private

[datquocnguyen]

We temporarily set the instruct version private to reinvestigate its safety level. We will make it public within this week.

[datquocnguyen]

Available, works with all transformers version.

[hacbachvotu]

Why do I get this error, author?

[datquocnguyen]

code?

[hacbachvotu]

code?

Here it is, thanks for your support!

[datquocnguyen]

does the following code work in your side?

from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer  

model_path = "vinai/PhoGPT-7B5-Instruct"  

config = AutoConfig.from_pretrained(model_path, trust_remote_code=True)  
config.init_device = "cuda"
# config.attn_config['attn_impl'] = 'triton' # Enable if "triton" installed!

model = AutoModelForCausalLM.from_pretrained(  
    model_path, config=config, torch_dtype=torch.bfloat16, trust_remote_code=True  
)
# If your GPU does not support bfloat16:
# model = AutoModelForCausalLM.from_pretrained(model_path, config=config, torch_dtype=torch.float16, trust_remote_code=True)
model.eval()  

tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)  

PROMPT = "### Câu hỏi:\n{instruction}\n\n### Trả lời:"  

input_prompt = PROMPT.format_map(  
    {"instruction": "xin chào"}  
)  

input_ids = tokenizer(input_prompt, return_tensors="pt")  

outputs = model.generate(  
    inputs=input_ids["input_ids"].to("cuda"),  
    attention_mask=input_ids["attention_mask"].to("cuda"),  
    do_sample=True,  
    temperature=1.0,  
    top_k=50,  
    top_p=0.9,  
    max_new_tokens=1024,  
    eos_token_id=tokenizer.eos_token_id,  
    pad_token_id=tokenizer.pad_token_id  
)  

response = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]  
response = response.split("### Trả lời:")[1]

[hacbachvotu]

does the following code work in your side?

from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer  

model_path = "vinai/PhoGPT-7B5-Instruct"  

config = AutoConfig.from_pretrained(model_path, trust_remote_code=True)  
config.init_device = "cuda"
# config.attn_config['attn_impl'] = 'triton' # Enable if "triton" installed!

model = AutoModelForCausalLM.from_pretrained(  
    model_path, config=config, torch_dtype=torch.bfloat16, trust_remote_code=True  
)
# If your GPU does not support bfloat16:
# model = AutoModelForCausalLM.from_pretrained(model_path, config=config, torch_dtype=torch.float16, trust_remote_code=True)
model.eval()  

tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)  

PROMPT = "### Câu hỏi:\n{instruction}\n\n### Trả lời:"  

input_prompt = PROMPT.format_map(  
    {"instruction": "xin chào"}  
)  

input_ids = tokenizer(input_prompt, return_tensors="pt")  

outputs = model.generate(  
    inputs=input_ids["input_ids"].to("cuda"),  
    attention_mask=input_ids["attention_mask"].to("cuda"),  
    do_sample=True,  
    temperature=1.0,  
    top_k=50,  
    top_p=0.9,  
    max_new_tokens=1024,  
    eos_token_id=tokenizer.eos_token_id,  
    pad_token_id=tokenizer.pad_token_id  
)  

response = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]  
response = response.split("### Trả lời:")[1]

I created a python file and used the code you sent, edited the path to the model in the internal drive and got this error. EDIT: I just imported "import torch" into the python file.