This model was fine-tuned for assisting pyschologists in assiting patients.
Use the code snippet below to load the base model and apply the adapter for inference:
1from transformers import AutoModelForCausalLM, AutoTokenizer
2from peft import PeftModel
3
4# Load the base model
5base_model_name = "trillionlabs/Trillion-7B-preview"
6adapter_path = "/path/to/adapter" # Replace with actual adapter path
7tokenizer = AutoTokenizer.from_pretrained(base_model_name)
8base_model = AutoModelForCausalLM.from_pretrained(base_model_name)
9
10# Apply the adapter
11model = PeftModel.from_pretrained(base_model, adapter_path)
12model = model.merge_and_unload()
13
14# Run inference
15input_text = "Your input text here"
16inputs = tokenizer(input_text, return_tensors="pt")
17outputs = model.generate(**inputs)
18print(tokenizer.decode(outputs[0], skip_special_tokens=True))
