HelpingAI-Lite/README.md

datasets, language, metrics, library_name, tags, license, widget, model-index

datasets	language	metrics	library_name	tags	license	widget	model-index
cerebras/SlimPajama-627B HuggingFaceH4/ultrachat_200k bigcode/starcoderdata HuggingFaceH4/ultrafeedback_binarized	en	accuracy speed	transformers	coder Text-Generation Transformers HelpingAI	mit	text <|system|> You are a chatbot who can code!</s> <|user|> Write me a function to search for OEvortex on youtube use Webbrowser .</s> <|assistant|> text <|system|> You are a chatbot who can be a teacher!</s> <|user|> Explain me working of AI .</s> <|assistant|>	name results HelpingAI-Lite task metrics type text-generation name type value Epoch Training Epoch 3 name type value Eval Logits/Chosen Evaluation Logits for Chosen Samples -2.707406759262085 name type value Eval Logits/Rejected Evaluation Logits for Rejected Samples -2.65652441978546 name type value Eval Logps/Chosen Evaluation Log-probabilities for Chosen Samples -370.129670421875 name type value Eval Logps/Rejected Evaluation Log-probabilities for Rejected Samples -296.073825390625 name type value Eval Loss Evaluation Loss 0.513750433921814 name type value Eval Rewards/Accuracies Evaluation Rewards and Accuracies 0.738095223903656 name type value Eval Rewards/Chosen Evaluation Rewards for Chosen Samples -0.0274422804903984 name type value Eval Rewards/Margins Evaluation Rewards Margins 1.008722543614307 name type value Eval Rewards/Rejected Evaluation Rewards for Rejected Samples -1.03616464138031 name type value Eval Runtime Evaluation Runtime 93.5908 name type value Eval Samples Number of Evaluation Samples 2000 name type value Eval Samples per Second Evaluation Samples per Second 21.37 name type value Eval Steps per Second Evaluation Steps per Second 0.673

HelpingAI-Lite

Subscribe to my YouTube channel

Subscribe

GGUF version here

HelpingAI-Lite is a lite version of the HelpingAI model that can assist with coding tasks. It's trained on a diverse range of datasets and fine-tuned to provide accurate and helpful responses.

License

This model is licensed under MIT.

Datasets

The model was trained on the following datasets:

• cerebras/SlimPajama-627B
• bigcode/starcoderdata
• HuggingFaceH4/ultrachat_200k
• HuggingFaceH4/ultrafeedback_binarized

Language

The model supports English language.

Usage

CPU and GPU code

from transformers import pipeline
from accelerate import Accelerator

# Initialize the accelerator
accelerator = Accelerator()

# Initialize the pipeline
pipe = pipeline("text-generation", model="OEvortex/HelpingAI-Lite", device=accelerator.device)

# Define the messages
messages = [
    {
        "role": "system",
        "content": "You are a chatbot who can help code!",
    },
    {
        "role": "user",
        "content": "Write me a function to calculate the first 10 digits of the fibonacci sequence in Python and print it out to the CLI.",
    },
]

# Prepare the prompt
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)

# Generate predictions
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)

# Print the generated text
print(outputs[0]["generated_text"])