Download the model
# This is to set the path to save the model
from pathlib import Path
models_path = Path.home().joinpath('Question_Generation_model', 'UTeMGPT')
models_path.mkdir(parents=True, exist_ok=True)
# Download the model
from huggingface_hub import snapshot_download
my_model = snapshot_download(repo_id="KLimaLima/finetuned-Question-Generation-mistral-7b-instruct", local_dir=models_path)
To load the model that have been downloaded
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.
from unsloth import FastLanguageModel
model, tokenizer = FastLanguageModel.from_pretrained(
model_name = my_model,
max_seq_length = max_seq_length,
dtype = dtype,
load_in_4bit = load_in_4bit,
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
This model uses alpaca prompt format such as below
alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
### Instruction:
{}
### Input:
{}
### Response:
{}"""
instruction = 'Write an inquisitive question about a specific text span in a given sentence such that the answer is not in the text.'
sentence = "I want to bake a cake during my free time. I need to know the ingredients that need to be use."
inputs = tokenizer(
[
alpaca_prompt.format(
instruction,
sentence,
"", # output - leave this blank for generation!
)
], return_tensors = "pt").to("cuda")
To generate output
outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
tokenizer.batch_decode(outputs)
Uploaded model
- Developed by: KLimaLima
- License: apache-2.0
- Finetuned from model : unsloth/mistral-7b-instruct-v0.2-bnb-4bit
This mistral model was trained 2x faster with Unsloth and Huggingface's TRL library.
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
HF Inference API was unable to determine this model’s pipeline type.
Model tree for KLimaLima/finetuned-Question-Generation-mistral-7b-instruct
Base model
unsloth/mistral-7b-instruct-v0.2-bnb-4bit