metadata
language:
- en
library_name: transformers
pipeline_tag: text-generation
widget:
- text: >-
Instruct: You are a chatbot, who is helping to curate datasets. Based on
the input paragraph as context generate as many mcq question as possible
without repeptition. You donot generate repetitive questions.
When you are given a paragraph for context. You will generate multiple mcq
questions, it's 4 options and it's actual answer.
For Example:
Paragraph: .....
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: b)....
-End of Question-
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: d)....
-End of Question-
and so on.
Paragraph: Computer science theories and basic programming principles form
the foundation of the ever-evolving field of technology. At its core,
computer science is not just about writing code but involves the
exploration and application of fundamental principles that underpin the
design and functioning of computers. One key theory in computer science is
the Turing Machine, proposed by Alan Turing in the 1930s. This theoretical
construct laid the groundwork for understanding the limits and
possibilities of computation. The idea that any computable function could
be computed by a Turing Machine provided a theoretical framework for the
development of modern computers. Another essential theory in computer
science is the concept of algorithms. Algorithms are step-by-step
procedures or formulas for solving problems and performing tasks. They are
crucial in programming as they guide the computer in executing tasks
efficiently. The study of algorithms involves analyzing their efficiency
and correctness, and it plays a pivotal role in designing software that
can handle large datasets and complex computations. Moreover, algorithms
are closely related to data structures, which are the ways in which data
is organized and stored in a computer's memory. Efficient data structures
are essential for optimizing the performance of algorithms.
Output:
example_title: Example 1
- text: >-
Instruct: You are a chatbot, who is helping to curate datasets. Based on
the input paragraph as context generate as many mcq question as possible
without repeptition. You donot generate repetitive questions.
When you are given a paragraph for context. You will generate multiple mcq
questions, it's 4 options and it's actual answer.
For Example:
Paragraph: .....
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: b)....
-End of Question-
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: d)....
-End of Question-
and so on.
Paragraph: Business financial education is an essential aspect of any
successful enterprise. It encompasses a range of knowledge and skills
necessary for effectively managing the financial aspects of a business,
including budgeting, financial analysis, investment strategies, and risk
management. A solid understanding of financial concepts enables business
owners and managers to make informed decisions that drive profitability
and sustainability. It empowers individuals within organizations to
interpret financial statements, assess performance metrics, and identify
opportunities for growth and improvement. Moreover, financial education
fosters accountability and transparency, ensuring that stakeholders have a
clear understanding of the financial health and trajectory of the
business. By investing in financial education, businesses can mitigate
risks, optimize resources, and ultimately achieve their long-term
objectives.
Output:
example_title: Example 2
- text: >-
Instruct: You are a chatbot, who is helping to curate datasets. Based on
the input paragraph as context generate as many mcq question as possible
without repeptition. You donot generate repetitive questions.
When you are given a paragraph for context. You will generate multiple mcq
questions, it's 4 options and it's actual answer.
For Example:
Paragraph: .....
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: b)....
-End of Question-
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: d)....
-End of Question-
and so on.
Paragraph: LLMs, or Language Model Models, are advanced artificial
intelligence systems designed to process and generate human-like text
based on input prompts. LLMs leverage sophisticated algorithms and vast
datasets to understand and generate coherent language across a wide range
of topics and contexts. Businesses and individuals can benefit from LLMs
in various ways, including content creation, customer support, language
translation, and data analysis. By leveraging LLMs, businesses can
automate repetitive tasks, streamline workflows, and improve efficiency.
Moreover, LLMs can assist in generating personalized content, enhancing
customer engagement, and driving conversions. To maximize the benefits of
LLMs, it's essential to understand their capabilities and limitations, as
well as best practices for integrating them into existing workflows.
Additionally, staying updated on advancements in LLM technology and
investing in ongoing training and development can ensure that businesses
harness the full potential of these powerful tools to achieve their
objectives.
Output:
example_title: Example 3
This model is for the module
Initial Knowledge Assessment Test Generation
Steps
- Data was gathered by:
- Downloading youtube playlists for each course from every category
- The videos were transcribed
- The text was fed to chatgpt via API, to formulate prompts n reponse pairs.
- 2.78 Billion parameter Phi2 model by Microsoft was finetuned on the curated data.
How to use the model?
Note the format of the prompt. Only change the text in the variable "paragraph". This is the text which acts as the context for the generated test./
# Use a huggingafce pipeline as a high-level helper
from transformers import pipeline
import torch
pipe = pipeline("text-generation",
model="SalehAhmad/Initial_Knowledge_Assessment_Test-Model-Phi2_3Epochs",
device_map='auto',
torch_dtype=torch.bfloat16,
max_new_tokens=1024)
paragraph = '''Computer science theories and basic programming principles form the foundation of the ever-evolving field of technology. At its core, computer science is not just about writing code but involves the exploration and application of fundamental principles that underpin the design and functioning of computers. One key theory in computer science is the Turing Machine, proposed by Alan Turing in the 1930s. This theoretical construct laid the groundwork for understanding the limits and possibilities of computation. The idea that any computable function could be computed by a Turing Machine provided a theoretical framework for the development of modern computers.
Another essential theory in computer science is the concept of algorithms. Algorithms are step-by-step procedures or formulas for solving problems and performing tasks. They are crucial in programming as they guide the computer in executing tasks efficiently. The study of algorithms involves analyzing their efficiency and correctness, and it plays a pivotal role in designing software that can handle large datasets and complex computations. Moreover, algorithms are closely related to data structures, which are the ways in which data is organized and stored in a computer's memory. Efficient data structures are essential for optimizing the performance of algorithms.'''
prompt = f'''Instruct: You are a chatbot, who is helping to curate datasets. Based on the input paragraph as context generate as many mcq question as possible without repeptition. You donot generate repetitive questions.
When you are given a paragraph for context. You will generate multiple mcq questions, it's 4 options and it's actual answer.
For Example:
Paragraph: .....
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: b)....
-End of Question-
-Start of Question-
Question: ......
Options:
a) .....
b) .....
c) .....
d) .....
Actual Answer: d)....
-End of Question-
and so on.
Paragraph: {paragraph}
Output: '''
output = pipe(prompt,
num_return_sequences=1,
return_full_text=False)
print(output[0]['generated_text'])