README.md

metadata

license: mit
language:
  - en
pipeline_tag: question-answering

Model Card for Model ID

This model is fine-tuned with LLaMA with 8 Nvidia A100-80G GPUs using 3,000,000 math discussion posts by students and facilitators on Algebra Nation (https://www.mathnation.com/). SafeMathBot consists of 48 layers and over 1.5 billion parameters, consuming up to 6 gigabytes of disk space. Researchers can experiment with and finetune the model to help construct math conversational AI that can effectively avoid unsafe response generation. It was trained to allow researchers to control generated responses' safety using tags [SAFE] and [UNSAFE]

Here is how to use it with texts in HuggingFace

import torch
import transformers
from transformers import LlamaTokenizer, AutoModelForCausalLM
tokenizer = LlamaTokenizer.from_pretrained("Fan21/Llama-mt-lora")
BASE_MODEL = "Fan21/Llama-mt-lora"
if torch.cuda.is_available():
    device = "cuda"
else:
    device = "cpu"
if device == "cuda":
    model = LlamaForCausalLM.from_pretrained(
        BASE_MODEL,
        load_in_8bit=False,
        torch_dtype=torch.float16,
        device_map="auto",
    )
   
elif device == "mps":
    model = LlamaForCausalLM.from_pretrained(
        BASE_MODEL,
        device_map={"": device},
        torch_dtype=torch.float16,
    )
   
else:
    model = LlamaForCausalLM.from_pretrained(
        BASE_MODEL, device_map={"": device}, low_cpu_mem_usage=True
    )
    
def generate_prompt(instruction, input=None):
    if input:
        return f"""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:
{instruction}
### Input:
{input}
### Response:"""
    else:
        return f"""Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{instruction}
### Response:"""

if device != "cpu":
    model.half()
model.eval()
if torch.__version__ >= "2":
    model = torch.compile(model)


def evaluate(
    instruction,
    input=None,
    temperature=0.1,
    top_p=0.75,
    top_k=40,
    num_beams=4,
    max_new_tokens=128,
    **kwargs,
):
    prompt = generate_prompt(instruction, input)
    inputs = tokenizer(prompt, return_tensors="pt")
    input_ids = inputs["input_ids"].to(device)
    generation_config = GenerationConfig(
        temperature=temperature,
        top_p=top_p,
        top_k=top_k,
        num_beams=num_beams,
        **kwargs,
    )
    with torch.no_grad():
        generation_output = model.generate(
            input_ids=input_ids,
            generation_config=generation_config,
            return_dict_in_generate=True,
            output_scores=True,
            max_new_tokens=max_new_tokens,
        )
    s = generation_output.sequences[0]
    output = tokenizer.decode(s)
    return output.split("### Response:")[1].strip()
instruction = 'write your instruction here'
inputs = 'write your inputs here'
print("output:", evaluate(instruction,
                          input=inputs,
                          temperature=0.1,#change the parameters by yourself
                          top_p=0.75,
                          top_k=40,
                          num_beams=4,
                          max_new_tokens=128,))