gmonsoon's picture
Update README.md
644a50f verified
metadata
widget:
  - messages:
      - role: system
        content: >-
          You are a science educator. The user will provide you with an
          individual looking for guidance in their daily education purposes, and
          your task is to assist them in determining what career path they are
          most suited for based on their skills and interests. You should also
          conduct research into the various options available, explain the job
          market trends in different industries, and advice on which
          qualifications would be beneficial for pursuing particular fields.
      - role: user
        content: Hello Sir!
      - role: assistant
        content: Hi, kids! How may I help you?
      - role: user
        content: >-
          I am interested in developing a career in natural science. What would
          you recommend me a book to read?
  - messages:
      - role: system
        content: You are a knowledgeable assistant. Help the user as much as you can.
      - role: user
        content: How to become smarter?
  - messages:
      - role: system
        content: You are a helpful assistant who provides concise responses.
      - role: user
        content: Hi, Sensei!
      - role: assistant
        content: Hello there! How may I help you?
      - role: user
        content: I need to solve this math problem. What the answer for this, 2x+8=18?
  - messages:
      - role: system
        content: >-
          You are a very creative researcher. User will give you a task, which
          you should complete with all your knowledge.
      - role: user
        content: Write the first paragraph about research in renewable energy.
inference:
  parameters:
    max_new_tokens: 256
    penalty_alpha: 0.5
    top_k: 4
license: apache-2.0
language:
  - en
pipeline_tag: text-generation
datasets:
  - Locutusque/hyperion-v2.0

This model is frankenmerge from phi-2. Model is expanded into 4b parameters and then, finetuned with Locutusque/hyperion-v2.0 (25k)

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "frankenmerger/delta-4b-instruct-v0.1"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])