Ebrahimaabdelghfar/Ubuntu_assistant_Gemma2B

Model Trained Using AutoTrain

This model was trained using AutoTrain. For more information, please visit AutoTrain.

Usage

  !pip install -U "huggingface_hub[cli]"
  !huggingface-cli login --token "************" --add-to-git-credential

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_path = "Ebrahimaabdelghfar/Ubuntu_assistant_Gemma2B"
torch.backends.cuda.enable_mem_efficient_sdp(False)
torch.backends.cuda.enable_flash_sdp(False)

tokenizer = AutoTokenizer.from_pretrained(model_path,max_new_tokens=1000000)
model = AutoModelForCausalLM.from_pretrained(
    model_path,
    device_map="auto",
    torch_dtype='auto'
).eval()

# Prompt content: "hi"
messages = [
    {"role": "user", "content": "what sudo do?"}
]

input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
output_ids = model.generate(input_ids.to('cuda'),max_length=1023,max_new_tokens=100)
response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)

print(response)