Spaces:
Runtime error
Runtime error
File size: 1,016 Bytes
46c8cfc b02a8ce d4b6d9f 46c8cfc 0710e36 46c8cfc d4b6d9f 46c8cfc d4b6d9f 46c8cfc d142b67 46c8cfc d4b6d9f 46c8cfc 0710e36 46c8cfc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 |
# Install necessary libraries
# !pip install transformers
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
import torch
# Model and prompt details
model_name = "mlabonne/llama-2-7b-guanaco"
prompt = "What is a large language model?"
# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.float16,
device_map="auto",
offload_folder="offload/folder" # Replace with the path to the offload folder
)
# Generate text using the provided prompt
sequences = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
device=0, # Change to the appropriate device index or "cuda" if using GPU
)(
f'<s>[INST] {prompt} [/INST]',
do_sample=True,
top_k=10,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
max_length=200,
)
# Print the generated text
for seq in sequences:
print(f"Generated Text: {seq['generated_text']}")
|