116 lines
5.1 KiB
Markdown
116 lines
5.1 KiB
Markdown
---
|
|
base_model: p-e-w/Llama-3.1-8B-Instruct-heretic
|
|
tags:
|
|
- text-generation-inference
|
|
- transformers
|
|
- unsloth
|
|
- llama
|
|
- Uncensored
|
|
- trl
|
|
- roleplay
|
|
- conversational
|
|
license: apache-2.0
|
|
language:
|
|
- en
|
|
pipeline_tag: text-generation
|
|
datasets:
|
|
- N-Bot-Int/Iris-Uncensored-Reformat-R2
|
|
- N-Bot-Int/RP-Mixed-v1
|
|
library_name: transformers
|
|
new_version: N-Bot-Int/OpenElla-NovelWriter-Requiem-Ascended
|
|
---
|
|
## NEW VERSION RELEASED!
|
|
**N-Bot-Int/OpenElla-NovelWriter-Requiem-Ascended** is officially released! our last and final novelWriter model!
|
|
you can visit the model card on [huggingface](https://huggingface.co/N-Bot-Int/OpenElla-NovelWriter-Requiem-Ascended)!
|
|
|
|
|
|
## // BORKED :/ - visit the other versions if ya want!
|
|
(The model exhibits covertion to uncensored RP and Godmods!)
|
|
- 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 `*`).
|
|
|
|
*READ MORE FOR MORE INFO*
|
|
8 BILLIONS PARAMS MODEL
|
|
|
|
# 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](mailto: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!
|
|
[](https://ko-fi.com/J3J61D8NHV)
|
|
|
|
Benchmarks are also in the pipeline and will be added once available.
|
|
|
|
---
|
|
|
|
# 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](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
|
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
|
- Fine-tuned Using:
|
|
- Kaggle Free Tier with T4 x2 |