Edit model card

Cerebras-GPT-Alpaca-590m

This model is a fine-tuned version of cerebras/Cerebras-GPT-590M on the tatsu-lab/alpaca dataset. This is a test to check if the concept of Databrick's databricks/dolly-v1-6b can be applied to smaller models.

Model description

This is an LLM trained to follow instructions, similar to ChatGPT, except much much smaller.

Sample Use

from transformers import pipeline

generator = pipeline('text-generation', model='linkanjarad/Cerebras-GPT-Alpaca-590m')

prompt = '''### Instruction:
Who is the first person on the moon?

### Response:'''

output = generator(prompt, do_sample=True, min_length=10, max_length=50, temperature=0.5)
print(output[0]['generated_text'])

Intended uses & limitations

The model works best when given input in the format of the dataset it was trained on. Here is an example of the format of the tatsu-lab/alpaca dataset:

### Instruction:
What are the three primary colors?

### Response:
The three primary colors are red, blue, and yellow.

Here are a few sample model outputs:

### Instruction:
List 5 characteristics of a good parent.

### Response:
A good parent should have the best communication skills, are able to manage the needs of their kids without fear of punishment, and are also motivated to make the best decisions possible. They should also be respectful of their children's needs, be able to understand their needs and provide support, and be willing to take on challenges. Furthermore, a good parent should possess the skills and intelligence to work with and be able to provide a healthy helping hand to children. Finally, a good parent should be willing to work with children, as it can be difficult to keep them engaged and motivated. Parents should strive to develop their children's skills, be able to make decisions and help them learn from them, and encourage them to pursue their goals. The good parent should also be a good role model, be able to lead their children to their best successes, and be willing to provide a supportive relationship. Finally, a good parent should
### Instruction:
Generate a new and original business plan for a product or service related to AI.

### Response:
The plan for the new product or service should include a comprehensive plan for AI-powered AI systems that will enable businesses to automate a wide variety of tasks such as making decisions, analyzing data, and creating product designs. The plan should include a detailed review of an innovative solution, and should also incorporate customer feedback and industry-specific requirements. The plan should also include a detailed review of the business case for an AI product or service. Additionally, the plan should include a detailed review of the company's competitive positioning, customer segmentation and customer feedback. Additionally, the plan should include a detailed review of product and pricing, as well as customer feedback and user research. Finally, the plan should involve both internal and external stakeholders, as well as the customer feedback, to ensure that stakeholders are satisfied with the plan. This plan should help ensure that the right solutions are available to
### Instruction:
Write a very short story about hamsters and whales.

### Response:
Once upon a time, there was a small hamster who lived on a beach in a remote village in the Pacific Ocean. One day, he decided to take a hike to the nearby town of San Francisco. He had heard about the whale and decided to learn more about it. He spent the day learning about the whale and its behavior, and spent the day visiting the local coral reef. He eventually arrived at the shore, where he saw the whales in their graceful and majestic forms. The journey would not be the same without the whale. He was astounded by what he had encountered the first time. He saw so much beauty and kindness, and it was a pleasure to see the beauty and kindness of the whale. He would never forget the day of the trip and knew that this gentle creature was his only source of sustenance. He would never forget the whales, who were so abundant,
### Instruction:
Explain why it's important to evaluate evidence and not just accept it at face value.

### Response:
Evaluating evidence, regardless of its type, can be a valuable way of understanding the issue. It helps us develop a better understanding of the issue, as well as help us make better decisions about our future. It also helps us to understand that our actions, decisions, and actions are often not just about the truth, but about the process of our lives. Evaluating evidence can help us stay up to date on current events and trends, while avoiding the distraction of unnecessary and unnecessary decisions. In other words, it helps us to learn from the past and make better decisions. As a result, it's important to evaluate evidence and not just accept it at face value. It's important to learn how to evaluate evidence through proper judgment, and to always evaluate evidence in a manner that is consistent with the evidence. Finally, it's important to look at the

Training Procedure

The model was trained with Huggingface's 🤗 Trainer for about 0.5 epochs.

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • gradient_accumulation_steps: 128
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.15
  • num_epochs: 1

Framework versions

  • Transformers 4.27.4
  • Pytorch 1.13.0
  • Datasets 2.1.0
  • Tokenizers 0.13.2
Downloads last month
17
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train linkanjarad/Cerebras-GPT-Alpaca-590m