NEW VERSION RELEASED!

// BORKED :/ - visit the other versions if ya want!

• if this is, what you want! Aka Novel formatting(get it.. Novel Writer!)... no seriously... A story generator
• Then this model is for YOU

OpenElla-NovelWriter-8B-V2 is OFFICIALLY RELEASED!

• IMAGE GENERATED USING CHATGPT!

OpenElla-NovelWriter-8B-V2, LIMITLESS EXPERIENCE, EXECELENT RP, EFFICIENT WEIGHT SIZE!

• Built on top of p-e-w/Llama-3.1-8B-Instruct-heretic, OpenElla-NovelWriter-8B-V2 is our NEWEST and MOST CAPABLE roleplay-focused model to date, trained with a clear goal: deliver the best open RP experience on accessible hardware, RIVALING OUR CHAMPION MISTHENA7B!.
• OpenElla specializes in Roleplay scenarios, with strong support for narrative writing, character immersion, and proper RP formatting (dialogue in " and actions in *).

OpenElla-NovelWriter-8B-V2 Model Procedure/Methodology:

• OpenElla-NovelWriter-8B-V2 is trained Using P-E-W-S's Llama-3.1-8B Instruct Version of the Model, ensuring full creative flow without refusal! OpenElla-NovelWriter-8B-V2 is then fed our brand new 17K DATASET named MEulysis-SYNTHETIC-M1!
• MEulysis-SYNTHETIC-M1 — a 17K entry synthetically generated dataset produced using multiple capable RP models (including Mythomax and Hermes), then carefully cleaned, formatted, and tuned across a wide range of roleplay scenarios.

What makes MEulysis-SYNTHETIC-M1 different?

• 17,000 entries of synthetic RP data generated from multiple frontier RP models
• Wide RP focus covering diverse character types, settings, and narrative styles
• Includes MOST ELABORATE EXPERIENCES to MOST TABOO SCENARIOS
• Clear formatting guidance baked into the data — proper use of "dialogue" and *actions*
• Removal of ANY FORMATTING, ensuring Dataset's Usability to other Architectures!
• Fully cleaned and curated for quality and consistency

Training Details

• Finetuning Tool: Unsloth AI + Huggingface TRL
• Training Platform: Kaggle Free Tier with T4 x2!
• Epochs: 3
• Final Training Loss: 0.8 (Lowest we've achieved so far!)
• Dataset: MEulysis-SYNTHETIC-M1 (17K entries)
• OpenElla-NovelWriter-8B-V2 is Our Brand New Powerful Model, If you ever encountered any issue, Want to commission us, or have any suggestions, please email us directly through nexus.networkinteractives@gmail.com we value any reports, suggestions to how we improve future Model, Once again feel free to finetune the model to your likings, However please consider Adding this Page for CREDITS
• Please handle the AI with Care and ethical considerations, when FINETUNING this AI model, due to its UNCENSORED Nature.
• We are not responsible for what this model generates. Use it responsibly and legally. You downloaded it, you own what you do with it.

What's Coming Next?

🔒 Human Fine-Tuned Version & Mixed Models — Coming Soon for Supporters First!
We're currently working on a Newer version of OpenElla to push the model's roleplay quality even further. Supporters will get early access before public release. Stay tuned!

Notices & Usage Tips

• Use Llama 3 format — the model is based on Llama 3, Soooooo using Llama 3 works best!.
• Calibrate per character card — every character is different, adjust your prompt, Model's settings(ie, temps, Top-K etc.) accordingly.

About

• OpenElla-NovelWriter-8B-V2 is
  • Developed by: N-Bot-Int
  • License: agpl 3.0
  • Finetuned from model : p-e-w/Llama-3.1-8B-Instruct-heretic

• Detail card:

  • Parameter
    • 8 Billion Parameters
    • (Please check your GPU Core, VRAM, CPU and RAM to see if you can comfortably run 8B models)
• Finetuning tool:
  • Unsloth AI
  • This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

    

  • Fine-tuned Using:
  • Kaggle Free Tier with T4 x2