Update model.py

model.py

@@ -3,11 +3,12 @@ str_cmd1 = 'pip install "unsloth[colab-new] @ git+https://github.com/unslothai/u
 str_cmd2 = 'pip install --no-deps xformers "trl<0.9.0" peft accelerate bitsandbytes'
 os.system(str_cmd1)
 os.system(str_cmd2)
-os.environ["CUDA_VISIBLE_DEVICES"] = "1"  # or "0,1" for multiple GPUs
+#os.environ["CUDA_VISIBLE_DEVICES"] = "0"  # or "0,1" for multiple GPUs
 
 
 from unsloth import FastLanguageModel
 import torch
+device = torch.device("cpu")
 max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally!
 dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
 load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.