PiVoT-0.1-Evil-a Open-source Text Generation Model - Multiple variants meet different text experiment needs

Pivot 0.1 Evil A

Developed by maywell

PiVoT-0.1-early is a model fine-tuned based on Mistral 7B, focusing on text generation tasks and having different variant versions to meet different experimental needs.

Large Language Model

Transformers

Supports Multiple Languages#Multilingual text generation #Evil tuned variants #Noise embedding enhancement

Downloads 478

Release Time : 11/26/2023

Model Overview

PiVoT is a fine-tuned model based on Mistral 7B, inheriting its advantages and performing well in text generation. It provides different variant versions, including specially adjusted ones.

Model Features

Fine-tuned based on Mistral 7B

Inherits the advantages of Mistral 7B and performs well in text generation tasks.

Multiple version variants

Provides models with different adjustment versions, including the Evil tuned version and the Noisy Embedding adjusted version, resulting in more diverse results.

Multilingual support

Supports text generation tasks in English and Korean.

Model Capabilities

Text generation

Multilingual text processing

Use Cases

Experimental research

Text generation experiment

Conduct text generation experiments using different variant versions to compare the diversity and quality of the generated results.

The Evil tuned version and the Noisy Embedding adjusted version can produce more diverse texts.

🚀 PiVoT-0.1-early

PiVoT-0.1-early is a finetuned model based on Mistral 7B, offering different variations with unique tuning methods for text generation.

🚀 Quick Start

This section provides an overview of the PiVoT-0.1-early model and its variations.

✨ Features

Finetuned on Mistral 7B: PiVoT is a finetuned model based on the Mistral 7B architecture, leveraging its capabilities for text generation.
Variations Available:
- PiVoT-0.1-Evil-a: An "Evil tuned" version, finetuned using a specific method.
- PiVot-0.1-Evil-b: Has Noisy Embedding tuned, resulting in more diverse output.
Specific Prompt Template: Uses the Alpaca-InstructOnly2 prompt template for interaction.

📚 Documentation

Model Details

PiVoT is a finetuned model based on Mistral 7B. It is a variation from Synatra v0.3 RP, which has demonstrated decent performance.

PiVoT-0.1-Evil-a is an Evil tuned version of PiVoT, finetuned by the method described above.

PiVot-0.1-Evil-b has Noisy Embedding tuned, which leads to more variety in the results.

image/png

Prompt Template

The model uses the Alpaca-InstructOnly2 prompt template:

### Instruction:
{prompt}

### Response:

Disclaimer

⚠️ Important Note

The AI model provided herein is intended for experimental purposes only. The creator of this model makes no representations or warranties of any kind, either express or implied, as to the model's accuracy, reliability, or suitability for any particular purpose. The creator shall not be held liable for any outcomes, decisions, or actions taken on the basis of the information generated by this model. Users of this model assume full responsibility for any consequences resulting from its use.

💡 Usage Tip

The OpenOrca Dataset was used when finetuning PiVoT variations. Arcalive Ai Chat Chan log 7k, ko_wikidata_QA, kyujinpy/OpenOrca-KO and other datasets were used on the base model.

Follow me on twitter: https://twitter.com/stablefluffy

Consider supporting me in making these models alone: https://www.buymeacoffee.com/mwell or with Runpod Credit Gift 💸

Contact me on Telegram: https://t.me/AlzarTakkarsen

📄 License

The model is licensed under CC BY-SA 4.0.

Property	Details
License	CC BY-SA 4.0
Datasets	maywell/ko_wikidata_QA, kyujinpy/OpenOrca-KO, Anthropic/hh-rlhf
Languages	en, ko
Pipeline Tag	text-generation
Tags	not-for-all-audiences

image/png

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご