RuntimeError: cutlassF: no kernel found to launch!

#5
by Manmax31 - opened

I get the error RuntimeError: cutlassF: no kernel found to launch!

Here is my code:
`base_model_id = "upstage/SOLAR-10.7B-Instruct-v1.0"

tokenizer = AutoTokenizer.from_pretrained(base_model_id, trust_remote_code=True)
tokenizer.pad_token = tokenizer.eos_token

model = AutoModelForCausalLM.from_pretrained(
base_model_id,
load_in_8bit=True,
device_map="auto",
trust_remote_code=True,
torch_dtype=torch.bfloat16,
)

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

output = pipe(
prompt,
max_new_tokens=1000,
do_sample=True,
temperature=0.1,
top_k=50,
top_p=0.95,
return_full_text=False,
)
text = output[0]["generated_text"]
print(text )`

Any suggestions?
I have 4 Tesla V100 GPUs

upstage org

Hmm, it seems we cannot reproduce this error. Do you still have the problem?

hunkim changed discussion status to closed

yess facing same error

Sign up or log in to comment