| --- |
| license: apache-2.0 |
| tags: |
| - unsloth |
| - Uncensored |
| - text-generation-inference |
| - transformers |
| - unsloth |
| - llama |
| - trl |
| - roleplay |
| - conversational |
| language: |
| - en |
| base_model: |
| - unsloth/Llama-3.2-1B-Instruct |
| pipeline_tag: text-generation |
| library_name: peft |
| metrics: |
| - character |
| - bleu |
| - rouge |
| --- |
| # This is the Next Experimental OpenElla Architecture, Tested and Proven to Outcompete Minimaid-L2's and L1's |
| If you loved our Models, then please consider donating and supporting us through Ko-fi! |
| [](https://ko-fi.com/J3J61D8NHV) |
| Our Training Capabilities is limited to Google Colab's Free tier So any Help, Donation or Renting is very much appreciated! |
| |
|
|
|  |
|
|
| # OpenElla StoryWriter - TypeB |
| - Introducing OpenElla StoryWriter - TypeB! |
| - OpenElla StoryWriter TypeB is our brand new Model trained on a 1B llama Instruct model, capable of writing good RP responses, |
| without sacrificing Speed or Performance! The New Architecture is trained on **HERETICAL** and finetuned for RP dataset. |
| Which paves the path for the model's incredible performance compared to other 1B models, and are capable of Generating responses |
| On-par to 3B counterparts |
|
|
| # OpenElla StoryWriter - TypeB Base-Model Card Procedure: |
| - **TypeB** |
| TypeB Model contains 1B params, Type A is yet to be released and will contain 3B params Counterpart. There is **NO PLAN** to bump the |
| training model to **7B**, this is due to Google colab's free tier limit, limiting our training time and the intensity of training! |
|
|
| TypeB is trained using **HERETICAL** tool, its base model, Llama3.2 1B was taken from HF, and was trained on Heretical to remove as much Wastes as much as possible. |
| However, please bear in mind that the model is **UNCENSORED** or atmost should be due to heretical training, |
|
|
| **N-BOT-INT is NOT LIABLE FOR ANY DAMAGES, MISUSE, OR UNINTENDED OUTCOMES RESULTING FROM USER-GENERATED PROMPT** |
|
|
| TypeB is then feeded 600K dataset aimed to increase the performance of the AI model in term of these categories |
| - ROMANCE |
| - MYSTERY / DETECTIVE |
| - SCI-FI |
|
|
| This, alongside the same 600K datasets Handwritten nature, gave the Model length and consistency which is Useful for RP purposes, |
| HOWEVER DUE TO LACK OF PARAMS, EXPECT THE MODEL TO HALLUCINATE |
|
|
| **ALL FINETUNING, ACCOUNTING DATASET PREPARATION ASWELL, TOOK 5 DAYS. HERETICAL TRAINING IS EXCLUDED BECAUSE THE HERETICAL HAS BEEN PREPPED LONG TIME AGO** |
|
|
| - OpenElla StoryWriter - TypeB is Another Instance of Our Smallest Model Yet! if you find any issue, then please don't hesitate to email us at: |
| [nexus.networkinteractives@gmail.com](mailto:nexus.networkinteractives@gmail.com) |
| about any overfitting, or improvements for the future Model, |
| Once again feel free to Modify the LORA to your likings, However please consider Adding this Page |
| for credits and if you'll increase its **Dataset**, then please handle it with care and ethical considerations |
|
|
| |
| - OpenElla StoryWriter - TypeB is |
| - **Developed by:** N-Bot-Int |
| - **License:** apache-2.0 |
|
|
| - MiniMaid-L3 Official Metric Score |
|  |
| "Minimaid Instances Vs. OpenElla StoryWriter - TypeB" |
| |
|  |
| "OpenElla StoryWriter - TypeB vs. Several Different AI Models" |
| |
| - **METRIC SCORES ARE HIGHLY SUBJECTIVE, Feel free to use the model and judge them yourselves!** |
|
|
| - # Notice |
| - **For a Good Experience, Please use** |
| - (PLEASE CALIBRATE THE MODEL DEPENDING ON THE CHARACTER CARD) |
| - USE **CHATML**, the AI MODEL IS FINETUNED TO USE CHATML more than any other format |
|
|
| - # Detail card: |
| - Parameter |
| - 1 Billion Parameters |
| - (Please check your GPU Core, CPU and RAM to see if you can confortably run 1B 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: |
| - Google Colab Free Tier |
