MeowGPT-ll3

MeowGPT Readme

Overview

MeowGPT, developed by CutyCat2000x, is a language model based on Llama with the checkpoint version ll3. This model is designed to generate text in a conversational manner and can be used for various natural language processing tasks.

Usage

Loading the Model

To use MeowGPT, you can load it via the transformers library in Python using the following code:

from transformers import LlamaTokenizer, AutoModelForCausalLM, AutoTokenizer

tokenizer = LlamaTokenizer.from_pretrained("cutycat2000x/MeowGPT-ll3")
model = AutoModelForCausalLM.from_pretrained("cutycat2000x/MeowGPT-ll3")

Example Prompt

An example of how to prompt the model for generating text:

{{ bos_token }}{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' '  + content.strip() + eos_token }}{% endif %}{% endfor %}

The <s> and </s> are start and end tokens.

About the Model

• Base Model: Llama3
• Checkpoint Version: ll3
• Datasets Used: Open-Orca/SlimOrca-Dedup, jondurbin/airoboros-3.2, microsoft/orca-math-word-problems-200k, m-a-p/Code-Feedback, MaziyarPanahi/WizardLM_evol_instruct_V2_196k, mlabonne/orpo-dpo-mix-40k

Citation

If you use MeowGPT in your research or projects, please consider citing CutyCat2000x.

Disclaimer

Please note that while MeowGPT is trained to assist in generating text based on given prompts, it may not always provide accurate or contextually appropriate responses. It's recommended to review and validate the generated content before usage in critical applications.

For more information or support, refer to the transformers library documentation or CutyCat2000x's resources.